In both Dutch and French, morphological and syntactic patterns are available to build multi-word expressions: for instance, Du. wetenschapsbeleid ‘science policy’ vs wetenschappelijk beleid ‘scientific policy’ (Booij 2019) or Fr. village(-)vacances vs village de vacances ‘holiday resort’ (Van Goethem & Amiot 2019). However, Dutch has a stronger tendency towards compounding than French (among others, Van Goethem 2009). Little attention has been paid so far to the impact of such cross-linguistic differences on the use of compounds in the L2 and in translation, even though word-formation awareness has been proven crucial for learners’ L2 proficiency and creativity (Balteiro 2011), and constitutes an important factor in producing target-like translations (Lefer 2012). Results from a previous study on learner data in the MulTINCo corpus from 195 learners in French-speaking Belgium (Meunier et al. 2020, Hendrikx & Van Goethem 2024) indicate that French-speaking learners of Dutch overuse phrasal structures in contexts where a compound should be used (e.g. lessen van zwembad ‘classes from swimming pool’ instead of zwemlessen ‘swimming classes’), as expected based on the cross-linguistic differences between French and Dutch. However, the corpus results also show that learners produce different types of creative compounds, such as compounds in cases where a simplex word is appropriate (e.g. kookman ‘cookman’ instead of kok ‘chef’) or codeswitching within compounds (e.g. verjaardagsgateau ‘birthday cake’). The translation data were newly collected in 2024 and include French-to-Dutch student translations from bachelor students (18-20 years), whose L1 is Dutch, at Dutch-speaking universities in Belgium. To increase the comparability of the learner and the translator data, the students were asked to translate the native French texts from MulTINCo into Dutch. Moreover, the translator data are compared to the L1 Dutch subcorpus of MulTINCo. Preliminary findings from the translation ...