John M. Levis

COMPUTER TECHNOLOGY IN TEACHING AND RESEARCHING PRONUNCIATION

Pronunciation, long on the periphery of applied linguistics research and pedagogy, continues to grow in importance because of its central roles in speech recognition, speech perception, and speaker identity. Pronunciation-related issues such as comprehensibility, accent, and the mutual intelligibility of varieties of world Englishes are central to many questions in applied linguistics. This calls for a sophisticated understanding of how technological tools that have long been used to shed light on phonological categories can be applied to teaching. Research into computer-assisted pronunciation teaching (CAPT) suggests that both researchers and pronunciation teachers increasingly make use of technology to answer key questions, to ensure that claims are defensible, and to develop theories and practices that more closely match acoustic reality. This article reviews three key areas where computer technology and pronunciation intersect: (1) appropriate pedagogical goals and the measurement of improvement; (2) the ability of CAPT to give useful, automatic feedback; and (3) the use of technology in diagnosing pronunciation errors. This article concludes with recommendations for key technological competencies needed by any researcher or teacher who examines pronunciation-related issues.

Pronunciation and the Assessment of Spoken Language

The ACTFL Guidelines for foreign language proficiency, the dominant measure for assessing foreign language proficiency in the United States, describe ten different levels of proficiency.1 Although they are extensive and detailed, the guidelines are strikingly random in describing how pronunciation contributes to speaking proficiency. Four of the levels (Advanced Mid, Advanced Low, Intermediate High, and Novice Mid) do not mention pronunciation, while three others (Intermediate Mid and Low, Novice High) suggest that pronunciation may be important as evidence of L1 influence, and thus, it appears, lower levels of proficiency. Another level, Advanced High, mentions only ‘precise vocabulary and intonation’ (Breiner-Sanders et al., 2000: 15). Only at the top level, Superior, and the bottom, Novice Low, do descriptors suggest a significant contribution for pronunciation. Superior speakers are able to use suprasegmentals to support coherent discourse, while Novice Low speakers may be unintelligible due to poor pronunciation. The overall effect of this haphazard collection of descriptors suggests that pronunciation is relatively unimportant in determining speaking proficiency. Evidence of L1 influence (common in almost all adult learners) automatically places speakers below the advanced level, and being intelligible does not appear to be closely related to pronunci-ation. Yet any experienced teacher knows that L2 speakers can be both very advanced in most elements of speaking ability and still be unintelligible due to pronunciation errors.

A project-based approach to teaching research writing to nonnative writers

It is commonly accepted that writing instruction should meet the specific needs of writers and that students in scientific and technical fields benefit more by learning to write to match the requirements of their specific fields. A variety of models for writing classes have been proposed to meet these needs, from genre-based approaches to courses targeting specific disciplines to general courses serving a heterogeneous group of students from many disciplines. Although persuasive arguments can be made for discipline-specific writing courses, many writing courses for nonnative writers at U.S. universities operate with two key constraints. First, monetary and curricular limitations mean that students from a variety of disciplines are placed in the same course. Second, these courses are staffed by instructors who, while well-prepared in addressing language needs of nonnative writers, may know very little about the content and conventions of engineering and science. This paper discusses a writing course which works within these constraints and has been developed for graduate students who are early in their program of study. In the course, groups of students carry out an original research project as a vehicle to learn professional writing conventions common to research papers in a variety of scientific and engineering fields. In addition, students analyze written conventions in published articles within their fields to raise awareness of how general conventions are worked out in their individual disciplines. General principles for the course are discussed, and samples of successful research topics are provided.

The Intonation and Meaning of Normal Yes/No Questions.

The typical intonation of yes/no questions in American English is usually thought to be high- rising. Accounts of yes/no questions also assume that the intonation of these questions is different from that of standard British English, which is thought to use a low-rising intonation. This article challenges this belief and reports on a study that shows that American speakers of English do not distinguish between the high-rising and low-rising intonation, and argues that the supposed difference in intonation between the varieties is not accurate for yes/no questions. Finally, the results indicate that intonation of yes/no questions should be an unimportant issue in English language teaching because intonation on these questions appears to play a minimal role in the success of interactions between speakers from different varieties of English.

Social dynamics in second language accent

This volume offers a definitive source for understanding social influences in L2 pronunciation, demonstrating the importance of empirical findings from a number of research perspectives, and outlining the directions that future work can take. The aim is to present a coherent argument for the significance of social factors and how they contribute to phonological acquisition.

Research into practice: How research appears in pronunciation teaching materials

Research into pronunciation has often disregarded its potential to inform pedagogy. This is due partly to the historical development of pronunciation teaching and research, but its effect is that there is often a mismatch between research and teaching. This paper looks at four areas in which the (mis)match is imperfect but in which a greater recognition of research can lead to better teaching materials (high variability phonetic training, intonation, information structure, and setting priorities). Furthermore, two areas in which teaching materials are desperate for research to be carried out (connected speech and the primacy of suprasegmentals) will be discussed.

Automatic Speech Recognition

Automatic speech recognition (ASR) is an independent, machine-based process of decoding and transcribing oral speech. A typical ASR system receives acoustic input from a speaker through a microphone, analyzes it using some pattern, model, or algorithm, and produces an output, usually in the form of a text (Lai, Karat, & Yankelovich, 2008). Keywords: CALL

Native speakers and native users: loss and gain

and the teachers’ constant individual and collective reflection on their practice. Touching on several topics which have been receiving increasing attention over the last decade in Europe – including multimodal pedagogy, CLIL, pluriliteracy, and the accommodation of newcomers in mainstream classrooms – this book poses several thought-provoking questions that have the potential to generate further research, both in Catalan-medium schools and in other historically minoritised language contexts. It is both relevant and engaging, and will appeal to scholars of multilingual education and second-language acquisition, as well as to adolescent students, language acquisition, as well as to students and researchers interested in the ways in which top-down language policies interact with grassroots efforts to adapt to changing sociolinguistic and sociocultural realities.

Pronunciation for English as an international language: from research to practice

Since Jennifer Jenkins (2000) wrote The Phonology of English as an International Language, pronunciation teachers and researchers have waited for a serious extension of her ideas. Jenkins’ approach, now called English as a Lingua Franca (ELF) but originally under the rubric of English as an International Language (EIL), can be viewed as complementary to the larger framework of World Englishes (WE), in which mutual intelligibility of different Englishes, including issues of pronunciation, was a central concern. WE and EIL, however, remained on the periphery of pronunciation teaching until Jenkins argued that pronunciation was directly relevant to intelligibility in the interactions between non-native speakers. Her book created an explosion of interest in pronunciation, with arguments about whether her Lingua Franca Core (LFC) was relevant to ESL/EFL pronunciation teaching. This new book is the first scholarly extension of ideas found in Jenkins (2000). It pursues multiple aims in relation to EIL and pronunciation teaching. First, it explicates WE, EIL and ELF, describing their similarities and differences. Second, it summarizes EIL research relevant to pronunciation in general and to specific features. Third, the book applies research findings to pronunciation teaching. Fourth, it incorporates acoustic phonetics findings in its discussion. The combination of these four areas makes the book unique.