Wêr stiet de te-ynfinityf yn it Frysk?

Abstract

West Frisian verbal clusters are characterised by a fairly strict complement-head ordering. However, so-called to-infinitives (preceded by te and ending in -n) show a different behaviour: some appear to the left of their selecting verb, some are realised at the the right-hand side of the rest of the verbs. So far, scholars agree that to-infinitives are a systematic exception to the basic verb ordering, this meaning that verbal to-infinitives are always on or to the right of the clustered verbs. Consequently, to-infinitives preceding the other verbs are analysed as non-verbal. Part of the long-running debate concentrates on the nature of those non-verbal infinitives. This debate, however, is based on an extremely limited set of data. This situation begs both an empirical and a theoretical question, namely firstly: are the different types of to-infinitives really in the positions where the scholars claim them to be? And secondly: how can the odd distribution of West Frisian to-infinitives best be explained? In this article, we will present the findings of a new data collection. These show that the literature has described the situation of the 19th century fairly well, but not so the Frisian spoken by people born through­out the 20th century, in which to-infinitives increasingly adjust to comple­ment-head ordering. We will try to account for this change by claiming that complement-head ordering in verb clusters knows no systematic exceptions and by showing that the observed changes are brought about by reanalysis of the relevant structures as compounds of a main verb and an auxiliary.