Beginner learners' progress in decoding L2 French: Transfer effects in typologically similar L1-L2 writing systems

Research into second language (L2) print-to-sound decoding has consistently underlined the importance of L1-to-L2 transfer. Facilitation has been reported where the L1 and L2 writing systems are typologically similar (e.g., English and French) rather than distant (e.g., English and Chinese). However, other studies have found that young beginner learners of French, whose L1 is English, have poor L2 decoding proficiency and make little progress in this area, despite the similarities between these writing systems. To explore this apparent contradiction, 76 young L1 English speakers learning French as their L2 completed a French decoding test at two time points: at the end of their first and third trimesters of learning French in secondary school. Rather than simply being judged correct or incorrect, as in previous studies, their oral productions were transcribed and analysed at the level of individual grapheme-phoneme correspondences. Participants were found to make very little progress in L2 decoding, as measured by the number of grapheme tokens realised acceptably. Detailed analysis of their realisations of two vowel graphemes further showed that their decoding was often consistent with L1 symbol-sound mappings; and that it appeared to involve orthographic units larger than the individual grapheme, as is required in order to decode some graphemes accurately in English. However, not all attested realisations were explainable on the basis of English symbol-sound mappings, perhaps because the L2 vowels sometimes occurred in unfamiliar spelling bodies which did not readily trigger L1 symbol-sound associations. Viewed from a cognitive processing perspective, these findings suggest that the similarities between the English and French writing systems—whilst being facilitative in some respects—may also pose particular difficulties for the L2 learner.