Orthographic Interference in the Acquisition of /h/ by L1 French Learners of English

While orthographic input has been shown to help in the acquisition of L2 phonological contrasts (Weber & Cutler, 2004; Escudero, 2008), more recent evidence suggests a negative effect of L1 transfer depending on the congruency of the grapheme-to-phoneme correspondence (GPC) between the L1 and the L2 (Hayes-Harb et al, 2010; Escudero et al, 2014; 2015). This study looks at the impact of orthography on the acquisition of the phoneme /h/ by Francophone L2 learners of English, who continue to make errors of deletion and insertion well into higher levels of proficiency. While the grapheme in French is consistently silent, its pronunciation in English is inconsistent: it is articulated only in onset position with some lexical and positionally-conditioned exceptions. The question addressed here is whether the presence or absence of the orthography facilitates or inhibits L1 French learners’ encoding of English /h/, and what is the effect of this inconsistency in the GPC of the L2. In a word-learning experiment, adult L1 French listeners were taught English pseudowords by associating auditorily presented stimuli with pictures, with or without the spelling. They were placed into three learning conditions: auditory + congruent spelling, auditory + congruent/incongruent (inconsistent) spelling, and auditory only. Later, they were tested on their memory of the words via an auditory word-picture matching task followed by a word reading task (Steele, 2002; Bassetti et al., 2015). In the word-picture matching task, no difference was found in accuracy rates at post-test between the auditory only and congruent spelling groups; however, rates for the congruent/incongruent group were significantly lower and at chance, suggesting a negative L1 orthographic interference when the GPC of the target language is inconsistent. Results for the word-reading task will be discussed along with implications for pronunciation teaching.