Morphological ability among monolingual and bilingual speakers in early childhood: The case of two Semitic languages

Abstract

This study examines the possible effects of bilingualism, mother tongue and type of morphology on morphological awareness of Arabic- and Hebrew-speaking preschoolers (mean age – 5:4). Four groups of children participated in the study: (1) 50 Arabic-speaking monolingual speakers; (2) 50 Hebrew-speaking monolingual speakers; (3) 50 Arabic/Hebrew bilingual speakers; and (4) 50 Hebrew/Arabic bilingual speakers. Participants from the bilingual groups were sequential non-balanced bilingual speakers who started learning a second language at ages 3–4 in a bilingual Arabic/Hebrew kindergarten. All children performed two tasks on inflectional morphology and three tasks on derivational morphology in one or both languages. To examine inflectional morphology, domain plural nouns were chosen because of their linear nature in both Hebrew and Arabic and because inflectional plural-noun morphology is acquired very early. In derivational morphology, the focus was on the verbs because of their high token frequency, early acquisition compared to nominal morphology, and its importance for Semitic languages. The results demonstrate significant effects of mother tongue, bilingualism and type of morphology on the children’s performance. The better results were obtained in Hebrew-speaking monolinguals and in Arabic-speaking bilinguals. Monolingual Hebrew speakers performed better in Hebrew than Arabic-speakers did in Arabic. At the same time, Arabic-speaking bilingual children demonstrated significantly better results in Hebrew (second language) than Hebrew speakers did in Arabic (second language). Analysis of the findings also shows that differences in performance among the bilingual and monolingual groups seem to relate not only to psycholinguistic factors such as linguistic complexity but also to sociolinguistic factors (e.g. diglossia in Arabic).

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 25 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.