THE EFFECTS OF BI-LITERACY ON PHONOLOGICAL AWARENESS AND READING ABILITY IN L2 ITALIAN AND L3 ENGLISH - A STUDY ON YOUNG HERITAGE SPEAKERS OF ROMANIAN IN ITALY

If it holds that L3 learners are influenced by the general cognitive effects of bilingualism (Bialystok, 2001) and that bilingualism provides better opportunities for the growth of cognitive, linguistic, and academic skills provided bilinguals develop literacy in their first language (Cummins, 2000), it could be hypothesised that bi-literacy translates into benefits not only in L2 but also in L3 learning. This dissertation aims at determining the effects of bi-literacy (versus bilingualism alone) on young bilingual children’s phonological awareness and reading ability. The study specifically looks for the possibility of metalinguistic awareness benefit on the level of phonology and word-level decoding by having Romanian acquired as a first and heritage language and Italian as a second language and medium of instruction. It also examines the possibility of transfer of these benefits into the third language, English, which is learnt formally at school as a foreign language. Finally, it seeks to determine whether this possible metalingual benefit can manifest differently in L2 with respect to L3. Sixty-one 8 to 10-year-old Romanian-Italian bilingual children who met the exclusionary criterion of having Romanian as first and dominant language in the home were selected from different public primary schools in Northern Italy that were hosting Romanian language courses for young heritage speakers. Next, they were further divided into two groups, namely the bi-literate bilingual group that were participating in these courses of Romanian as a heritage language and therefore had literacy skills in Romanian, and the mono-literate bilingual group that was not attending any such course nor had literacy knowledge in Romanian. Finally, a group of Italian monolingual children, matched on demographic and socio-economic characteristics, were selected from the same schools as the bilingual groups. All participants were first assessed for their non-verbal intelligence. Then bilingual children’s language proficiency was measured through three vocabulary tests, each in Romanian, Italian and English, and were assessed for their daily (passive and active) spoken language use in Romanian and Italian through a parental questionnaire. They were also assessed for their phonological awareness abilities that included phoneme segmentation and blending, syllable blending and onset-rime oddity in the three languages. Finally, the participants were assessed for their Italian and English word reading abilities through a list of both real and pseudo-words arranged in increasing difficulty level. The monolingual children completed the vocabulary, phonological awareness and reading tests in Italian and English only. Results from the above tests generated four important findings. Firstly, it is clear that the bi-literate bilinguals develop better phonological awareness skills in Romanian, their L1, as they outperformed the mono-literate bilinguals on all phonological awareness tasks. Secondly, the superiority of the bi-literates over mono-literate bilinguals is also visible in L2 Italian, at least as pseudo-word reading, phoneme segmentation, onset-rime oddity and syllable blending are concerned. Moreover, being Romanian-Italian bi-literate can also translate into benefits over Italian monolinguals as tests on L2 Italian onset-rime, syllable and pseudo-words have shown; at the same time, as far as L2 Italian testing is concerned, being mono-literate bilinguals does not result in advantages over monolinguals. Thirdly, the results supported the prediction that Romanian (L1) literacy skills would have a positive impact on English (L3) phonological awareness and literacy acquisition. This conclusion derives from the superiority of the bi-literate bilingual group over both the mono-literate bilingual and monolingual groups on all four phonological awareness tests and on the word decoding task. On the other hand, being bilingual mono-literates results in leverage over monolinguals only on phoneme and onset-rime testing. Finally, these results would seem to put forward that L3 acquisition is different from L2 acquisition and cannot be considered a simple variation of it, since the development of phonological and reading skills resulted to manifest themselves differently in the two languages. Therefore, it could be suggested that L2 and L3 acquisition should represent a distinct research domain. This study contributes to early heritage language literacy learning in the Italian and European context, especially learning to read in English as a foreign language. My study provides evidence that among bilingual children, the phonological processing skill and reading ability skills can be elevated through teaching the orthographic knowledge of the first language.