1

When Frequency Is NotEnough: Explaining Bilingual

Children’s MorphologicalAcquisition

Johanne Paradis, University of AlbertaElena Nicoladis, University of AlbertaMartha Crago, Université de Montréal

Child Language Seminar, Newcastle upon Tyne, July 2006

Usage-Based Theory and BilingualAcquisition

• Input structure and frequency key mechanismsunderlying acquisition patterns (Tomasello, 2003)

• Simultaneous bilinguals have their input spacedivided between two languages, so lessexposure to each language

• UB Theory predicts that bilingual childrenwould lag behind monolinguals in achievingacquisition milestones (Tomasello, 2004)

2

Bilingual Acquisition ofMorphosyntax

• School-age children:-language-at-home & language-at-school input

measures impact on bilingual children’s rate ofmorphosyntactic acquisition (e.g. Gathercole, 2002; 2006)

• Preschool/kindergarten children:– Conflicting findings on whether bilinguals

always lag behind monolinguals (Erdo et al, 2005;Gathercole & Thomas, 2005; Marchman et al, 2004; Nicoladis et al, inpress; Paradis & Genesee, 1996; Paradis et al, 2003)

Critical Mass, Complexity and Morphology

• Ceiling in performance (unlike vocabulary), so bilingualsmay “catch up” faster to monolinguals

• Critical mass of input needed varies depending oncomplexity of morphology to be acquired– Transparent vs. opaque

• Bilingual-monolingual differences are in rate, not sequencesof acquisition

§ Among bilingual homes, does dominant language make adifference?

§ How is transparency/opacity defined?

(N. Ellis, 2002; Gathercole, 2002, 2006; Gathercole & Hoff, in press)

3

Defining Opacity/Transparency:

• Bybee’s exemplar-based model of thelexicon - inflectional morphology

• Past tense forms in French and English

Exemplar-Based Model of theLexicon

• Multi-morphemic words stored fully inflected andinter-connected by

– phonological form– Semantic features

• Token frequency in input and output = increaseslexical strength of stem and stem +morphemeconstructions

Bybee (1995, 2001, 2002)

4

Exemplar-Based Model of theLexicon

• Type frequency (number of unique stem+morphemeconstructions in lexicon) increases schema strength

– Schema = rules like [verb [-ed]] = past tense reference– Type frequency = critical mass for productive and accurate use of

inflection• Irregular forms = inflectional islands

– Sensitive to token frequency in becoming established– Subject to overregularization due to superior strength of regular

schema

Bybee (1995, 2001, 2002)

5

Past Tense in French and English• English simple past

– regular [-ed] and irregular strong verbshe walks / he walked; she takes / she took / *she taked

• French passé composé– avoir/être + past participle– 1st conjugation: “regular” (based on type frequency)

marcher: Il marche / Il a marché (er = é)– 2nd & 3rd conjugation: families of “irregulars”

prendre: elle prend / elle a pris / *elle a prendu / *elle a prennéouvrir: Il ouvre / il a ouvert / *il a ouvri / *il a ouvré

Exemplar-Based Model andthe Past Tense in French and English• Transparent morphology = high type frequency of

schema• Irregular verbs are more opaque than regular

verbs - fewer types for each pattern/unique typesfor some patterns

• Irregulars would be later-acquired than regulars• Irregulars particularly vulnerable in case of

reduced input

NB: Words & Rules similar predictions for regulars and irregulars

6

Acquisition of the Past Tensein English and French

• Regular past tense– >90% correct at 4;6-4;11 in English (Rice & Wexler, 2001)

– >90% correct at 4;0-6;0 in French (Jakubowicz & Nash, 2001;Paradis & Crago, 2001)

• Irregular verbs– Accuracy with irregulars as a group lags behind regular

verbs in both English and French (Rice & Wexler, 2001;Nicoladis et al, in press)

– Overregularization errors found in both English andFrench (Marchman & Bates, 1994; Marcus et al, 1992; Nicoladis et al, inpress)

Do bilinguals lag behind monolingualsin their acquisition of the past tense?

1. Difference between bilinguals and monolingualssmaller for bilinguals’ dominant language

2. Difference between bilinguals and monolingualssmaller for regular than irregular past tense

3. No difference in bilinguals and monolinguals inacquisition sequences• regulars >> irregulars• overregularization errors

7

Participants• 23 French-English bilingual children aged

4;0-5;5 (simultaneous and early sequential)• 6 French monolingual children same age

range (more to come…)• Children residing in Edmonton, Canada• All children were attending French-language

preschool or kindergarten

Procedures• Parental questionnaire on input patterns• Peabody Picture Vocabulary Test (PPVT-III:

Dunn & Dunn, 1997)

• Échelle de vocabulaire en image Peabody(EVIP: Dunn et al, 1993)

• Past tense probe from the Test of EarlyGrammatical Impairment (TEGI: Rice & Wexler, 2001)

• Passé composé probe (experimenter-made)

8

Past tense probe: TEGI

“Here, the boy is painting. Now he is done. Tellme what he did”

Passé Composé Probe

“Camille vend du lait aux élèves dans sa classe.Maintenant elle a fini. Dis-moi ce qu’elle a fait.”

Camille is selling milk to the pupils in her class. Now,she’s finished. Tell me what she did.

9

Language Dominance• Dominant language = language for which child

receives more input• Measures

– simultaneous versus early sequential– rating scales of use of that language in the home by

each parent to the child– EVIP and PPVT z scores: verify categorization; break a

tie for two children• Children grouped as French or English dominant

English criterion-referencing(TEGI)

• 61% of all bilinguals scored at or abovethe age-expected criterion score fortypical language development on pastprobe as a whole (regular and irregularcombined)

• 87% of English-dominant bilingualsscores at or above criterion for probe asa whole

10

11

Irregular finite = irregular verbs correct + “overregularized”

2X3 mixed ANOVA produced significant main effects for language group andpast tense type, but no significant interaction. Regulars > irregulars. Frenchdominant bilinguals = monolinguals for regulars and irregulars

12

Irregular finite = irregular verbs correct + “overregularized”

Discussion and Implications• Bilinguals show the same acquisition patterns as

monolinguals overall– regulars > irregulars in English and in French– overregularization errors in English and French

• Bilinguals = monolinguals in their dominant language forregular past tense & irregular past finite

• Bilingual-all/monolingual differences smaller in olderchildren in English

• Bilinguals more difficulties with irregular verbs in Englishthan in French– English irregular verbs = greatest differences between

bilinguals and monolinguals

13

Discussion and Implications• Bilingual morphosyntactic acquisition is vulnerable

to these children’s reduced and variable input(predicted by UB Theory) - but not global delay

• Interacting factors important - no bilingual-monolingual differences for transparentmorphology

• School readiness/language assessment: languagedominance of child and transparency/opacity oftarget structure being probed are crucial to keep inmind

Many thanks to research assistants Aimée Berubé, HeatherGolberg, and Tamara Sorenson

This research was funded by the Alberta Heritage Foundationfor Medical Research and by the Social Sciences and

Humanities Research Council of Canada

johanne.paradis@ualberta.cahttp://www.ualberta.ca/~jparadis/

14

Selected ReferencesBybee, J. (1995). Regular morphology and the lexicon. Language and Cognitive Processes, 10, 425-455.Bybee, J. (2001). Phonology and Language Use. Cambridge: Cambridge University Press.Bybee (2002). Phonological evidence for exemplar storage of multiword sequences. SSLA, 24, 215-221.Ellis, N. (2002). Frequency effects in language processing. SSLA, 24, 143-188.Gathercole, V. (2002b).Monolingual and bilingual acquisition: Learning different treatments of that-trace

phenomena in English and Spanish. In D. K. Oller & R. Eilers (eds.), Language and literacy in bilingualchildren (pp.220-254). Clevendon: Multilingual Matters.

Gathercole, V. (2006). Miami and North Wales, so far and yet so near: Morphosyntactic development in bilingualchildren. Manuscript.

Gathercole, V., & Thomas, E. M. (2005). Minority language survival: Input factors influencing the acquisition ofWelsh. In J. Cohen, K.T. McAlister, K. Rolstad, and J. MacSwan (Eds.) ISB4: Proceedings (pp. 852-874).Somerville, MA: Cascadilla Press.

Gathercole, V. & Hoff, E. (in press). Input and the acquisition of language: Three questions. In E. Hoff & M.Shatz (Eds.), The Handbook of Language Development, Blackwell Publishers.

Jakubowicz, C. & Nash, L. (2001). Functional categories and syntactic operations in (Ab)normal languageacquisition. Brain and Language, 77, 321-339.

Marchman, V., & Bates, E. (1994). Continuity in lexical and morphological development: A test of the criticalmass hypothesis. Journal of Child Language, 21, 339-366.

Marchman, V., Martínez-Sussman, C. and Dale, P. (2004). The language-specific nature of grammaticaldevelopment: Evidence from bilingual language learners, Developmental Science, 7, 212-224.

Marcus, G., Pinker, S., Ullman, M., Hollander, M., Rosen, T.J., and Xu, Fei (1992). Overregularization inlanguage acquisition. Monographs of the Society for Resesarch in Child Development, 57(4).

Selected ReferencesNicoladis, E., Palmer, A. & Marentette, P. (in press). The importance of a critical mass of past tense verbs in

using past tense morphemes correctly. Developmental Science.Paradis, J., Crago, M., Genesee, F. & Rice, M. (2003). Bilingual children with specific language impairment:

How do they compare with their monolingual peers? Journal of Speech, Language and HearingResearch,46, 1-15.

Paradis, J. & Crago, M. (2001). The morphosyntax of Specific Language Impairment in French: Evidence for anExtended Optional Default Account. Language Acquisition, 9(4), 269-300.

Paradis, J. & Genesee, F. (1996). Syntactic acquisition in bilingual children: autonomous or interdependent?Studies in Second Language Acquisition, 18, 1-25.

Paradis, J., Crago, M. & Genesee, F. (in press). Domain-specific versus domain-generaltheories of the deficit in SLI: Object pronoun acquisition by French-English bilingual children. Language

Acquisition.Rice, M. & Wexler, K. (2001). Test of Early Grammatical Development: Examiner’s Manual. San Antonio, TX:

The Psychological Corporation.Tomasello, M. (2003). Constructing a language: A usage-based theory of language acquisition. Cambridge, MA:

Harvard University Press.Tomasello (2004). Lunchtime Debate: Where does language come from? 29th Annual Boston University

Conference on Language Development, Boston, MA.