In an online production experiment, we investigated the effect of sentence position on the preference for either a nominative or object form of an object pronoun restricted by a relative clause in Dutch. Results show a significant preference for the nominative form of the restricted object pronoun in sentence-initial position as it was chosen in 95% of the cases. In the original object position this percentage is only 20%. The preference for a nominative pronominal object is considered a grammatical norm violation. We account for this in terms of a combination of two factors. First, the presence of the relative clause makes the object ‘long’. Second, the sentence-initial position is a syntactic position that is relatively far removed from the original object position. We argue that when a long object is topicalized, there are too many intervening elements between the pronoun and the verb of which it is the complement. If the distance between the pronominal object and the verb has become too long, the object case fades from the working memory. This then results in the appearance of nominative case as the default case for topicalized object pronominal relative clauses in Dutch.

The current study investigates whether a Long Short-Term Memory (LSTM) network can learn the wh-island constraint in Dutch in a way comparable to human native speakers. After establishing with an acceptability judgement task that native speakers demonstrate a clear sensitivity to wh-island violations, the LSTM network was tested on the same sentences. Contrary to the results of the native speakers, the network was not able to recognize wh-islands and to block gap expectancies within them. This suggests that input and the network’s inductive biases alone might not be enough to learn about syntactic island constraints, and that built-in language knowledge or abilities might be necessary.

While kinship relations in Dutch are usually introduced by a possessive determiner, Twitter users have recently observed to use a definite article in that position. To learn more about the characteristics of this construction, we performed an exploratory investigation of the definite article possession construction with Dutch kinship terms on Twitter. We analysed 100 tweets for 24 kinship terms each, and annotated for the type of pre-nominal modifier used. Results show that the phenomenon is far from peripheral, as 13.2% of all selected tweets featured a definite article. The construction was most frequent with descending and horizontal relationship terms, and with improper kin terms (i.e., terms with a non-kin meaning at least as prominent as kinship use; Dahl & Koptsjevkaja-Tamm 2001: 202). These findings were explained by pointing to redundancy and the comical effect of distancing the construction creates.