Journal of Second Language Pronunciation - Volume 6, Issue 3, 2020

Abstract

This special issue revisits an extraordinarily influential paper for L2 pronunciation research and teaching (Munro & Derwing, 1995) by looking again at the original paper with new eyes and new analyses. The special issue also includes invited papers addressing current approaches based on the three key constructs in Munro and Derwing (1995): Intelligibility, comprehensibility, and accentedness. Papers in the issue include a reconsideration of the Intelligibility and Nativeness principles from Levis (2005), applications of the constructs to L2 Spanish (Nagle & Huensch), a consideration of how everyday L2 use affects comprehensibility (Zielinski & Pryor), long-term effects of intensive instruction (French, Gagné & Collins), influences on listener reaction to L2 speech (Isaacs & Thomson), empirical evidence for the dynamic nature of comprehensibility (Trofimovich et al.), a study on ELF intelligibility and functional load considerations (Thir), the relationship between comprehensibility and social evaluation of speech (Vaughn & Whitty), and one book review.

Abstract

We revisit Munro and Derwing (1995a), providing retrospective commentary on our original methods and findings. Using what are now well-established assessment techniques, the study examined the interrelationships among accentedness, comprehensibility, and intelligibility in the speech of second-language learners. The key finding was that the dimensions at issue are related, but partially independent. Of particular note was our observation that speech can be heavily accented but highly intelligible. To provide a fresh perspective on the original data we report a few new analyses, including more up-to-date statistical modeling. Throughout the original text we intersperse insights we have gained over the past 25 years. We conclude with retrospective interpretations, including thoughts on the relevance of the study to contemporary second language teaching and especially pronunciation instruction.

Abstract

Levis (2005) named two conflicting approaches to pronunciation teaching, the Nativeness Principle and the Intelligibility Principle. This paper revisits those two principles to argue for the superiority of the Intelligibility Principle in regard to where pronunciation fits within the wider field of language teaching, in how it effectively addresses teaching goals, in how it best addresses all contexts of L2 pronunciation learning, and in how it recognizes the reality of social consequences of pronunciation differences. In contrast, the Nativeness Principle, despite its long pedigree and many defenders, falls short by advocating native pronunciation for L2 learners, which is both unlikely to be achieved and unnecessary for effective communication in the L2.

Abstract

This study investigated relationships among intelligibility, comprehensibility, and accentedness in the speech of L2 learners of Spanish who completed a prompted response speaking task. Thirty native Spanish listeners from Spain were recruited through Amazon Mechanical Turk to transcribe and rate extracted utterances, which were also coded for grammatical and phonemic errors, and speaking rate. Descriptively, although most utterances were intelligible, their comprehensibility and accentedness varied substantially. Mixed-effects modeling showed that comprehensibility was significantly associated with intelligibility whereas accentedness was not. Additionally, phonemic and grammatical errors were significant predictors of intelligibility and comprehensibility, but only phonemic errors were significantly related to accentedness. Overall, phonemic errors displayed a stronger negative association with the listener-based dimensions than grammatical errors. These findings suggest that English-speaking learners of Spanish are not as uniformly intelligible and comprehensible as FL instructors might believe and shed light on relationships among speech constructs in an L2 other than English.

Abstract

In this longitudinal study we tracked change in comprehensibility and English use over a 10-month period in 14 L2 English learners (8 beginner, 6 intermediate) settling in Australia. They were interviewed 4 times during the 10 months as part of a larger longitudinal study. English use was reported at each interview using a language map and excerpts from recordings of Interviews 1 and 4 were rated for comprehensibility. Intermediate participants tended to be more comprehensible and maintain a higher level of English use over time than the beginners. Exploration of individual variation revealed a range of comprehensibility outcomes, the variable and non-linear nature of English use trajectories, and a possible relationship between comprehensibility change and English use for some participants. Important methodological implications for future studies relate to the measurement of comprehensibility and English use, the speech samples used for comprehensibility ratings, and the importance of individual variation.

Abstract

We assessed the long-term effects of intensive instruction on different aspects of L2 oral production. Adopting the tridimensional model of oral production (Munro & Derwing, 1995a), we compared high school learners who had received intensive ESL instruction (N = 42) with non-intensive learners (N = 39) on perceptual measures of L2 fluency, comprehensibility, and accentedness 4 years after a 5-month intensive instruction period. After controlling for academic ability and L2 proficiency, listeners’ ratings of fluency and comprehensibility were significantly higher for the IG; however, there was no specific group advantage for accentedness, suggesting both groups exhibited similar L2 accents. This study provides new empirical evidence that the oral fluency and comprehensibility benefits of an intensive experience may be long lasting, even when learners’ subsequent classroom exposure to the language is much more limited.

Abstract

This study investigates how Mandarin and Slavic language speakers’ comprehensibility, accentedness, and fluency ratings, as assigned by experienced teacher-raters and novice raters, align with discrete linguistic measures, and raters’ accounts of influences on their scoring. In addition to examining mean ratings in relation to rater experience and speaker first language background, we correlated ratings with segmental, prosodic, and temporal measures. Introspective reports were segmented, coded, enumerated, and submitted to loglinear analysis to elucidate influences on ratings. Results showed that ratings were strongly correlated with prosodic goodness and moderately correlated with segmental errors, implying the importance of both segmentals and prosody in L2 speech ratings. Experienced teacher-raters provided lengthier reports than novice raters, producing more comments for all coded categories where an error was identified except for pausing (a dysfluency marker). This may be because novice raters observed little else about the speech or struggled to pinpoint or articulate other features.

Abstract

This study examined longitudinal changes in second language (L2) interlocutors’ mutual comprehensibility ratings (perceived ease of understanding speech), targeting comprehensibility as a dynamic, time-varying, interaction-centered construct. In a repeated-measures, within-participants design, 20 pairs of L2 English university students from different language backgrounds engaged in three collaborative and interactive tasks over 17 minutes, rating their partner’s comprehensibility at 2–3 minute intervals using 100-millimeter scales (seven ratings per interlocutor). Mutual comprehensibility ratings followed a U-shaped function over time, with comprehensibility (initially perceived to be high) being affected by task complexity but then reaching high levels by the end of the interaction. The interlocutors’ ratings also became more similar to each other early on and remained aligned throughout the interaction. These findings demonstrate the dynamic nature of comprehensibility between L2 interlocutors and suggest the need for L2 comprehensibility research to account for the effects of interaction, task, and time on comprehensibility measurements.

Abstract

The Lingua Franca Core (LFC) proposes that NURSE is the only vowel quality important for international intelligibility, yet research findings regarding this issue are mixed. Moreover, it is unclear whether phonetic (rather than phonemic) substitutions of NURSE also affect international intelligibility more negatively than other phonemic vowel substitutions, though this seems unlikely on the basis of considerations of functional load (FL). This study compares the international intelligibility of two vowel substitutions typical of Austrian learners of English: the phonetic replacement of NURSE with a rounded and diphthongized vowel, and the phonemic replacement of TRAP with a vowel close to cardinal [e]. The findings suggest that, contrary to the LFC but in line with FL considerations, the phonetic substitution of NURSE is more intelligible to an international audience than the substitution of TRAP with [e]. However, differences in intelligibility between the two substitutions were largely ‘neutralized’ once contextual support was available.

Abstract

The processing fluency hypothesis proposes that listeners’ perceived difficulty processing the speech of L2 speakers (called comprehensibility/processing fluency) leads them to downgrade those speakers socially. In this paper, we investigate this relationship, focusing on context-specificity. L1-English listeners provided comprehensibility and social evaluation ratings of L1-Korean speakers speaking English, while an orthographic depiction of the speech either appeared alongside the audio or did not, a manipulation aiming to affect comprehensibility. Varying orthography between subjects, Experiment 1 found that orthography resulted in greater comprehensibility, but not more positive social evaluations. Experiment 2 manipulated orthography within subjects, varying context: orthography trials were presented first or last. Comprehensibility and social evaluation ratings were related only when orthography was first, suggesting a conditional, asymmetrical relationship where listeners more readily downgrade than upgrade the same speaker when orthography changes. Our results highlight the context-dependent nature of these constructs, limiting the generalizability of the processing fluency hypothesis.

This article reviews Intelligibility, oral communication, and the teaching of pronunciation9781108404013$34.99