Pronunciation in Progress (Pro2F): French Schwa and Liaison

Pronunciation training rarely takes place in current foreign language teaching, yet it is important for various reasons: Not only does good pronunciation help you to be understood, but also it allows you to be taken seriously and perceived as friendly. Nonetheless, for a long time foreign language pronunciation has been considered unlearnable. Recent research contests this long-standing view, but the necessary fundamental research for new pronunciation didactics still is lacking. The aim of the Pronunciation in Progress project (Pro2F; cf. French prof(esseur) `teacher`) is to fill this gap in the domain of French and to investigate on the basis of a new corpus in what manner Austrian pupils progress in acquiring schwa and liaison. In both cases, a sound is pronounced in certain contexts, but not in others, e.g. s(e)ra with or without schwa, les [z] amis with, but les copains without liaison. Both phenomena could easily be integrated into foreign language teaching, because they have more in common with grammar and vocabulary than with the classic pronunciation problems. Schwa and liaison behaviour of French native speakers has been the subject of extensive research, and the findings have influenced theory construction in phonology beyond the field of French Studies. In comparison, only some smaller, basically descriptive studies have been conducted on learners, mainly on students with English as L1, a high proficiency level in French and a stay abroad. These studies show that while acquiring the alternation between word forms with and without a vowel or consonant, errors can fossilize in certain contexts. A further challenge is the dominance of the standard language in the classroom (written texts and formal styles), which leads learners to overrealize schwas and liaisons. Only with intensive contact with spoken French, they develop a sensitivity for the stylistic and lexical variation. The Pro2F-corpus is innovative in three aspects: in terms of pupil-participants, it includes (1) the majority of the learners, (2) natives of Austrian German and (3) all learner levels from 0 to B1. It investigates more than 100 pupils from a Viennese secondary school, who have French as second foreign language and are in their first to sixth year of learning (ages 12 to 18). Following the methodology from the researcher network (Inter-) Phonologie du Français Contemporain (I)PFC, which has created a corpus of more than 600 native speakers of French, we compare reading aloud with spontaneous speech and learners with native speakers. On the basis of a quantitative analysis, we sketch a learning path for pronunciation from beginner to B1 competence level. The results flow back to the pupils via individual feedback for the participants, activities with the partner school, teacher development courses and a website with a popular science section. For the scientific community, we are providing a unique corpus, which can be the basis for further development of the Common European Framework for Languages and an improvement of pronunciation didactics.

The FWF project "Pronunciation in Progress: French Schwa and Liaison" Pro2F has furnished the biggest learner corpus of French worldwide, the first large corpus with data from high school students as the most important group of learners and the first corpus with L1 speakers of Austrian German. For this purpose, we recorded 145 students at a Viennese high school ranging from one to six years of learning (between 12 and 18 years old) reading word lists and texts, repeating a word list and in conversation with a French native speaker. The students of the 1st year have been recorded for the 6 subsequent years (2017/18-2023), in view of future longitudinal studies. The analyses show that the students know and master only some of the basic rules of French schwa and liaison: For example, they understand that the at the end of the word remains mute, and that following articles like "les" or pronouns like "nous", the liaison is nearly always realized. In the other contexts, the error rates are very high, which can be explained by the fact that they are not mentioned in the students' schoolbooks. This is also the case in contexts in which the liaison or the (non-)realization of schwa is mandatory and in which false realizations constitute grammatical errors and lead to comprehension problems. The sociolinguistic variation of French pronunciation is unknown to the students, even in the advanced years of learning. In addition, we observed numerous cases of negative transfer from the L1 Austrian German and the first foreign language English as well as an important influence of orthography on pronunciation. The new hypotheses developed in the project are: (1) Learners of French as a foreign language (FLE) only realize schwa, schwa deletion and liaison in contexts that are explicitly taught to them. (2) The erroneous quality of schwa and liaison consonants can in part be explained by the L1 grapheme-phoneme correspondences (e.g., realization of [e] instead of schwa or intermediate variants in the first syllable) and in part by the L1 phonological system (e.g., [s] instead of [z] or intermediate variants in liaison contexts). (3) FLE learners realize fewer liaisons in reading aloud than in spontaneous speech, in contrast to L1 speakers, due to little fluidity in this unusual task. Thus, the project has shown that good French pronunciation doesn't emerge on its own but must be explicitly taught and learned. In the cases of schwa and liaison, learners not only need to know the contexts in which the vowel or consonant has to/can be pronounced or remain mute but they also must master its correct quality.