The plural suffix -en (noot+en, 'nuts') is pronounced differently by speakers coming from different regions of the Netherlands. In this study, we compared the pronunciation of the plural suffix -en in phrases (noot+en kraken, 'to crack nuts') with linking en in compounds (noot+en+kroker, 'nutcracker'), because some claim that both are similar (Schreuder, Neijt, van der Weide, & Baayen, 1998), whereas others claim that they are not (Verkuyl, 2007). The pronunciations of 109 participants coming from five regions of the Netherlands were therefore compared in a picture naming task. A systematic relation between the pronunciations of plural -en and linking en was detected: Speakers from the Northern and Eastern regions produced [(upside-down e)n] most often for both the linking elements and plural endings, while speakers from the Middle and Western regions produced [upside-down e] most often for both. For speakers from the Southern region, we found no preference to pronounce either [upside-down e] or [upside-down e n] in compounds or phrases. It is concluded that Dutch speakers often do not distinguish plural -en from linking en in their speech production. Possibly, speakers of Dutch consider linking en and plural -en as the same morpheme.

No abstract

No abstract

No abstract

Abstarct An important skill for L1 language teachers when teaching grammar is the ability to produce and quickly evaluate arguments that underpin a grammatical analysis. Previous research has revealed that the strongest arguments in favour of a particular grammatical analysis are based on linguistic manipulations (LM) rather than on rules of thumb (RoT). This makes it critical for teachers to be able to handle arguments based on LM. If LM are considered too difficult to process compared to RoT, however, (student) teachers may avoid grammatical argumentation based on LM altogether, and they might struggle to evaluate their pupils’ LM-based grammatical argumentation. The current study has therefore examined whether LM impose a higher cognitive load on Dutch student teachers than RoT, using grammatical discussion tasks in which participants (N = 298) evaluated arguments based on RoT and on LM. Multilevel analyses indicate that LM are indeed more difficult to process than RoT, as measured by response times, correct classifications, and perceived difficulty ratings. This effect is partly influenced by student teachers’ need for cognition and their willingness to engage in grammar.