The intonation contour of non-finality revisited: implications for ESL teaching

International audiencePrevious research shows that English prosody is difficult to master for L2 learners of English (Grosser 1993, Gut 2009, amongst others) and this for different reasons (attitudinal function, type of discourse, context, etc.). Herment et al. (2014) show that one of the main difficulties for French ESL learners does not concern the form of the nuclear contour. They show that French speakers of English have a tendency to assign a rising pitch movement at the end of prosodic words, which leads to a clear difference in rhythm as compared to natives. It is all the more difficult for learners that it is generally admitted that an incomplete statement will be realized with a rising contour. According to Wells (2006), non-finality is signalled by a non- fall, i.e. a rise or a fall-rise. Cruttenden (2014) reports that the falling tone is the most common in British English (50%), followed by the rising tone and the fall-rise (40%). However, he notes that in reading, rises and fall-rises are more frequently used to indicate that a sentence is not finished. We tested these assumptions on the ANGLISH corpus (Tortel 2008), which is a collection of read speech, repeated sentences and monologues recorded by 20 native English and 40 French learners of English divided into 2 groups (beginners and advanced). We examined the productions of 10 native English female speakers and compared two speaking styles: read speech and spontaneous monologues. In total, 40 read short stories (200 sentences, statements) and 10 monologues were analysed, which represents 8.52 min of read speech and about 23 min. of natural speech. For this study, the theoretical framework chosen is the British school of intonation, following the idea that a configurational approach is better adapted to teaching (see Herment 2018). Gussenhoven (1983, 1984)’s tri-tonal approach is followed: fall (F), rise (R) and fall-rise (FR) are distinguished. For native speakers, the results for the reading task show that the most common tone for statements is the falling tone with 70,95% of falls, followed by 14,79% of rises and 13,97% of fall-rises. Concerning non-final statements, we found that out of 142 tone units, 78% were produced with a fall. Contrary to what is generally admitted, the rising tone is not the most common contour for incomplete statements as only 22% were found. It is to be noted that rising tones are much more frequent in spontaneous speech, which induces that the type of speech has to be taken into account. Our findings on the read speech have important pedagogical implications for French ESL learners. The teacher should encourage them to produce falling patterns in non-final statements, so as to help them avoid realizing those typically French rising contours on small units. More generally, this should constitute a guideline for teaching English intonation