Language Centres at a Crossroads

Dutch Universities require a CEFR English B2 level for admission to an English mediated (EMI) study programme. Previous research points out that B2 proficiency may diverge from the level that university staff actually expects (Meijer et al. 2012, Deygers, 2017), and does not match with what is required to understand and produce complex academic texts. Björkmann (2013) concluded that the formal quality of spoken English did not cause overt communicative disturbances in spoken academic settings, because speakers actively deploy pragmatic strategies. We present a study on written English as Lingua Franca (ELF) where students do not have the opportunity to negotiate or add meaning in interaction with the readers of their text. Our database is a set of written answers to open exam questions for an EMI bachelor in psychology. The exams were written in English by German students in the first year of study. Each written answer was previously graded by a course lecturer. These grades are available in the database. The data were gathered by De Vos (2019).

We carefully selected 45 text fragments with a length of between 150 and 250 words, with grades between 2 (the lowest score) and 10 (the max.). Four teachers in Academic English (language experts) added comments to the texts, on form features, comprehensibility and coherence, and assigned a CEFR level. The result is a list of characteristics of written ELF of German tertiary students. We make a distinction between features that do and do not cause disturbance. There was not always consensus on which features are typical of a particular CEFR level, nor on the CEFR levels of the text (ranging between B1 and C2; ICC = .783). In addition, we found no correlation between the assigned CEFR levels and the grades assigned for content at all. Informed content interpretation by the content expert seems to outweigh actual semantic coherence, overruling grammatical and lexical non-native characteristics.

These outcomes trigger the discussion on how to describe the higher CEFR levels beyond the language neutral can-do-statements in terms of linguistic competences for various languages, in our case for English (Hulstijn 2015).

Wir präsentieren die Forschungsergebnisse zu den akademischen Schreibfertigkeiten von deutschen Studenten im ESL, insbesondere ihre grammatikalische Korrektheit, die Verwendung des akademischen Vokabulars und die Textkohärenz. Diese Aspekte wurden von vier Sprachexperten, die Fachleute im Bereich des ESL-Unterrichts sind, auf der Grundlage einer schriftlichen Prüfungsaufgabe bewertet. Sie wiesen auch die GERS-Niveaus zu, damit wir die Nützlichkeit der GERS-Niveaus untersuchen konnten. Unser Ziel ist es letztendlich, herauszufinden, welche Arten des nicht muttersprachlichen Gebrauchs der englischen Sprache zu offensichtlichen kommunikativen Störungen führen oder in ELF-Situationen als nachteilig empfunden werden (Björkman, 2013). Unsere Ergebnisse weisen darauf hin, dass viele deutsche ESL- und EFL-Merkmale die Verständlichkeit eines Textes für inhaltsorientierte Prüfer nicht beeinflussen. Eine informierte semantische Interpretation überwiegt den tatsächlichen semantischen Koherenz, wobei grammatikalische und lexikalische, nicht-muttersprachliche Merkmale außer Acht gelassen werden.

References:

• Björkman, B. (2013). English as an academic lingua franca: An investigation of form and communicative effectiveness (Vol. 3). Walter de Gruyter.
• Deygers, B. (2017). Assessing high-stakes assumptions. A longitudinal mixed-methods study of university entrance language tests, and of the policy that relies on them. Herent, Belgium, ACCO Drukkerij.
• De Vos, Johanna F. (2019). Naturalistic word learning in a second language. Nijmegen: Radboud University dissertation.
• Hulstijn, J. H. (2015). Language proficiency in native and non-native speakers: Theory and research (Vol. 41). John Benjamins Publishing Company.
• Meyer, S., Gekeler, P., Manger, S., & Urank, D. (2013). Plurilingualism, multilingualism and internationalisation in the European Higher Education Area: Challenges and perspectives at a Swiss university. Language Learning in Higher Education, 2(2), 405-425.

Lidy Zijlmans studied Dutch Language and Literature and General Linguistics with an emphasis on language acquisition. She has been active in the field of Dutch as a Second Language (DSL) since 1978. Currently she is a senior associate for DSL at Radboud in'to Languages, the language centre of Radboud University Nijmegen. Previously she worked at the NT2 department of the VU Amsterdam and the James Boswell Institute at Utrecht University. She is a teacher, a developer of teaching tools, and a lecturer and publisher in teaching methodology for DSL. She taught DSL at the University of Duisburg-Essen and was a guest lecturer at the Erasmus Language Centre in Jakarta. Currently she is doing research on the role of second language in higher education. She presented results of this research at Cercles 2016.

Marc van Oostendorp (1967, PhD Tilburg University 1995) is Professor of Dutch and Academic Communication at Radboud University.

Prof. Roeland van Hout is emeritus professor in applied linguistics and variationist linguistics at the Centre for Language Studies of the Radboud University Nijmegen. He publishes in the fields of sociolinguistics, dialectology and second language acquisition and has a special interest in research methodology and statistics.