Held at the Association canadienne de linguistique appliquée/Canadian Association of Applied Linguistics Conference, Ottawa, Canada; 27 May 2009.
Over the past few decades perspectives on second language (L2) pronunciation have evolved from pessimistic appraisals of the capabilities of L2 learners and doubts about the value of instruction to a view of pronunciation teaching as an effective and important part of language pedagogy. Earlier research on the teaching of pronunciation dwelt extensively on the identification of learners' errors (mainly consonants and vowels) through comparative analyses. Until recently, little had been established about the effectiveness of pronunciation teaching, and pedagogical techniques were based more on speculation and theoretical notions than on empirically well-justified principles. More recent work addresses a broader range of issues relating not only to L2 phonological acquisition, but to the social implications of speaking with an accent and engaging with interlocutors, both native and non-native speakers of the L2. Among these are the relationship between accent and intelligibility, cognitive processes underlying phonological learning, the evaluation of L2 speech using impressionistic and acoustic techniques, prosodic influences on perception of accented speech, the role of ethnic affiliation and identity in L2 speakers' oral production, and the identification of misguided applications of knowledge about pronunciation by businesses and governments. These lines of work, along with empirical investigations of pronunciation instruction, engender a more sophisticated view of L2 phonological learning and teaching. Though further research remains to be done, important achievements have been made in identifying reasonable, achievable goals in the pronunciation classroom, establishing appropriate instructional foci, and evaluating outcomes. The presenters in this colloquium highlighted the major achievements of recent years and identified some of the important problems that remain.
Themes
Tracey Derwing (University of Alberta) & Murray Munro's (Simon Fraser University) presentation outlined the relationships among intelligibility (how much a listener actually understands), comprehensibility (effort required to understand) and accentedness (perception of difference from local variety), noting the partial independence of these dimensions. Intelligibility and comprehensibility are more closely related to each other than either is to accentedness. The former two serve as much more suitable goals than the reduction or elimination of accent, following the intelligibility principle (Levis Reference Levis2005). When guided by this principle, pronunciation instruction is effective, even in long-time L2 speakers.
Talia Isaacs (McGill University) further explored this tripartite distinction, emphasizing the need for a clearer understanding of these constructs to develop better tools for assessing both learner needs and progress in L2 pronunciation. Judgments of pronunciation are influenced by a variety of factors, including properties of the utterances themselves, listener biases, and context. A satisfactory evaluation of pronunciation is an essential part of language testing.
Mary Grantham O'Brien (University of Calgary) focused on L2 intonation, an understudied area of research whose importance is now recognized. In her research, English learners of German attempted to produce L2 intonation patterns; however, their over-compensation led to less native-like productions than would have been expected from a direct transfer of first language intonation. She identified the merits and limitations of visualization software in the teaching of prosody.
Elizabeth Gatbonton, Pavel Trofimovich & Norman Segalowitz (Concordia University) discussed the complexities of the relationship between ethnic group affiliation (EGA) and degree of accentedness in a study of the production of English interdental fricatives in adult francophones in Quebec. Their chief finding was the mediating role of L2 use in the relationship between EGA and pronunciation accuracy.
Helen Fraser's (University of New England) presentation emphasized that the sensitivity reliably seen in accent detection and rating tasks does not carry over to accent identification. Thus, the relatively poor ability of listeners to state what accent they have heard necessitates caution in language analysis to determine the origin of asylum seekers (LADO). She questioned the lack of transparency in the methods of some businesses contracted by governments for LADO purposes. Using a refugee appeal case, she pointed out how over-simplistic LADO analyses could endanger the lives of legitimate asylum seekers. She stressed the need for the involvement of qualified forensic linguists in LADO and for more research.
John Levis (Iowa State University) offered an overview of the current status of pronunciation research and teaching, indicating that L2 students and their interlocutors understand the importance of pronunciation in communication. Although pronunciation has been a neglected aspect of second language acquisition until recently, there are now conferences devoted to it, increasing numbers of books and scholarly articles on the topic, and strong professional interest groups. He noted that the current marketing of pronunciation to learners is based on fear, embarrassment and conformity, and proposed a new approach, focusing on the benefits of teaching pronunciation for enhanced communication. Such an approach would target L2 students with genuine intelligibility problems, practicing teachers, preservice teachers, teacher trainers and applied linguists.
The researchers concurred that pronunciation instruction is important, and that research must be extended to encompass a range of learning and socio-cultural contexts, ethical linguistic analysis, appropriate assessment, and cognitive processes.
