Learners’ pronunciations of familiar and unfamiliar French words: what can they tell us about phonological decoding in an L2?

Phonological decoding, defined here as converting the written forms of words (or letter strings) into the phonological forms they represent, has been argued to play an important role in various aspects of L2 learning. Previous studies have emphasized the importance of transfer in L2 decoding, interpretable as the automatic triggering of L1-based symbol-sound correspondences by L2 written input. However, in the case of familiar words, it may be possible for learners to retrieve pre-stored phonological representations based on the recognition of the words as graphic wholes, and thus to pronounce them correctly despite a lack of knowledge of symbol-sound correspondences at a sub-lexical level. To explore these issues, 31 below-intermediate learners of French in an English secondary school completed (a) a Reading Aloud Test, to assess their accuracy of decoding familiar and unfamiliar French words; (b) a scaffolded Picture Naming Task, to elicit their stored phonological pronunciations of the same familiar words; and (c) a vocabulary questionnaire, to assess their knowledge of the items included in the preceding two tests. Participants’ decoding in the Reading Aloud Test was more accurate for familiar than unfamiliar words, though still inaccurate in a majority of cases. Their pronunciations of the same familiar words on the Reading Aloud Test and Picture Naming Task sometimes differed. Further, their pronunciations of familiar items on the Picture Naming Task themselves showed some influence of English symbol-sound correspondences, despite the absence of written stimuli. Theoretical and pedagogical implications of these findings are discussed