Definite objects in the wild: A converging evidence approach to scrambling in the Dutch middle-field

Dutch allows a certain degree of freedom in the placement of definite objects with respect to that of adverbs in the middle-field of the clause. This phenomenon is known as scrambling. Dutch scrambling has been studied extensively over the past decades, yet the incentives to scramble definite objects are still matter of debate.

The general consensus is that scrambling of definite objects is motivated by discourse packaging conditions, in that presuppositional (topical and/or anaphoric) objects scramble and non-presuppositional (focused and/or non-anaphoric) objects do not. Most linguists are quite resolute in their claims about the obligatoriness of scrambling, but not everyone agrees with this strict “discourse template”. Empirical data (other than researcher intuitions) to back up these claims are scarce, and studies that do report empirical data do not corroborate the claims. Another factor argued to impact scrambling is the type of adverb. In particular, the structural position and focus sensitivity of adverbs should affect definite object placement.

This thesis shows through a series of experiments that the type of adverb plays a key role in the scrambling preferences of Dutch speakers, and that Dutch scrambling is at best influenced, but certainly not determined, by information structure. In addition, it charts the information structural partitioning of Dutch clauses since the 13th century, which serves as the basis for a theoretical analysis of scrambling. The results reveal striking discrepancies between expert judgments, folk judgments, and scrambling preferences in constrained language production. The thesis therefore advocates a “converging evidence” approach to Dutch scrambling.