First language effects on the identification and evaluation of second language speech

In Belgium, Dutch as spoken by Francophone learners is relatively frequent in political, commercial or educational contexts. While the characteristics of this second language (L2) variety have been studied extensively, there is to date no systematic report of how it is evaluated by either native speakers of Dutch or non-natives. Previous studies conducted in other language contexts have found that non-natives tend to be very critical towards L2 accents similar to their own. The main goal of the present study is to investigate the extent to which the listener’s first language (L1) impacts ratings of the fluency, accentedness and comprehensibility of L2 Dutch as spoken by Francophone learners and how it impacts the identification of the speakers’ L1. Specifically, we compared ratings by three groups of listeners: Francophone learners of Dutch, native speakers of Belgian Dutch and native speakers of Netherlandic Dutch. Moreover, the extent to which three additional cognitive and environmental factors influence L2 ratings is examined: listeners’ familiarity with the L2 variety, their language aptitude and language proficiency. The results show that the majority of native and non-native listeners recognized the speakers’ L1 (French). Non-native listeners perceived L2 speech as less fluent, less comprehensible and more accented than natives did, which corroborates the previously reported critical attitudes towards a shared L2 accent. Moreover, subtle differences in accent and fluency ratings were found between the Netherlandic Dutch and the Belgian Dutch listeners. No clear effects of other cognitive and environmental factors appeared in the ratings.

The contribution of prosody to intelligibility, comprehensibility and accentedness in foreign language acquisition

The purpose of this paper is to review recent research illustrating the importance of prosody in foreign language acquisition, and to examine whether music might help in this learning process. The paper starts off by defining prosody and by examining previous research on foreign language prosody acquisition, stating the difficulties for learners and the potential effect of non-native prosody on communication, notably on comprehensibility, intelligibility and accentedness. A subsequent section focuses on prosodic characteristics of Dutch and the problems foreign language learners may encounter in acquiring them. Based on this general description of foreign language prosody acquisition, the paper then zooms in on the link between music and prosody, and on the potential effect of musical training, musical abilities or the use of music in the foreign language classroom on foreign language prosody acquisition. The paper ends with a short discussion on avenues for future research.

Verstaanbaarheid als interactioneel en contextgebonden fenomeen

As a result of growing mobility and migration flows, the number of non-native speakers of Dutch in Belgium and the Netherlands have gradually increased over the past decades and so have the number of people enrolled in Dutch as a Second Language education. While there is huge variation in the profiles of these non-native speakers, they almost exclusively have in common that their Dutch sounds, in some way and at some stage, accented. In line with worldwide trends in foreign language teaching, the pronunciation goal in Dutch as a Second Language education has shifted from native-like to intelligible. Indeed, the notion of intelligibility has become prominent in language teaching and assessment. In this paper, we discuss the complexity of this notion and set it off against related terms like ‘comprehensibility’ and ‘foreign accent’. Through a literature review, we argue that intelligibility is an interactional and context-sensitive phenomenon: it is as much a responsibility of the speaker as it is of the listener or conversational partner(s) in general, whose attitudes will have an impact on the intelligibility and thus on the conversational flow and communicative success. After reviewing literature on the intelligibility of Dutch as a Second Language, we end by formulating some promising lines for future research.

Hebben we oren naar anderstaligen in het Vlaamse hoger onderwijs?

In order to enroll in Flemish higher education, aspiring students whose native language is not Dutch require a B2 CEFR level for language proficiency for most courses that are taught in Dutch. The question we aim to answer in this study is how future interlocutors of these students perceive that B2 entrance level. Do fellow students, lecturers and other members of staff feel that this level of language proficiency is sufficient to successfully navigate higher education? We specifically focus on the perception of pronunciation proficiency. We conducted a speaker evaluation experiment in which 191 students and staff members evaluated six sound clips with non-native Dutch speech representing pronunciation levels B1, B2 and C1. Results show that overall attitudes towards the speakers in the experiment are rather positive. In addition, there is a clear association between pronunciation proficiency and perceived functioning in higher education that reflects the Flemish higher education entrance policy. Our study also shows that various groups within the higher education community react differently towards non-native Dutch speakers and that there is a link between perception of non-native pronunciation and the amount of contact a listener has with non-native speakers. Overall, this study underlines the importance of pronunciation training in foreign language teaching for academic purposes.

Begrijpelijkheid en accent bij beginnende NT2-leerders

In social interaction, foreign language accent and comprehensibility impact how we perceive our conversational partners. In recent years, research interest in these constructs has been on the rise, while many issues remain underexposed. These issues include the relationship between comprehension and accent on the one hand, and background variables of both learner and assessor on the other. Since most research to date has been conducted with highly educated and advanced learners of English as a second or foreign language, we do not know to what extent those results can be generalised to a wider population that includes beginning learners of Dutch as a second language from various educational backgrounds. In addition, little research has been done into the comparability of the judgements of trained and non-trained assessors. In the current study, we compared the judgments of four trained evaluators with the intuitive judgments of 272 non-trained evaluators (first-year students at various Educational Bachelor’s Programmes in Primary Education). The first group of raters evaluated the speaking performance of 116 learners of Dutch as a second language using standardised criteria, the second group used more subjective criteria. The results show that the two groups of evaluators make very similar judgements and that these judgements are mainly related to two background variables: the nature of the NT2 course the learners followed (also an indicator of prior education and cognitive ability) and the level of language proficiency of the course.

Dutch listeners’ evaluations of degrees of German-accented Dutch

The effect of non-native accented speech on speaker and message evaluations is well-researched for English, but accent-induced effects in languages other than English have been underexplored. The purpose of the present study is to investigate the effects of a weak and strong German accent in Netherlandic Dutch on Dutch listeners’ perceptions of and attitudes towards the speaker and the message (in terms of, for instance, comprehensibility, persuasiveness, status, likeability and dynamism). In an experiment, 188 listeners evaluated fragments recorded in Standard Dutch guises, weakly German-accented guises and strongly German-accented guises. Findings showed that strongly German-accented guises were evaluated as less comprehensible and attracted less positive evaluations than weakly German-accented guises, whereas guises with weaker German accents were evaluated similarly to Standard Dutch guises. Comprehensibility mediated the effect of accent strength on speaker evaluations, which, in turn, mediated the effect of accent strength on persuasiveness. Thus, our findings highlight the importance of comprehensibility in determining speaker evaluations.