Taal en Tongval - Volume 64, Issue 1, 2012

Een typische eigenschap van het Nederlands (en andere West-Germaanse talen zoals het Duits en het Fries) is dat werkwoorden zich ophopen aan het einde van de zin in een zogenaamde werkwoordcluster. In hoofdzinnen bestaat die werkwoordcluster uit alle niet-finiete werkwoorden van de zin en in bijzinnen uit alle niet-finiete werkwoorden samen met het finiete werkwoord. De onderlinge volgorde van de werkwoorden in die werkwoordclusters vertoont veel variatie in het Nederlands. Die volgordevariatie staat centraal in dit themanummer. Deze inleiding geeft een stand van zaken van het onderzoek.

Gegenstand der vorliegenden Untersuchung ist die Verwendung unterschiedlicher Reihenfolge in verbalen Endgruppen im Mittelniederländischen des 13. und 14. Jahrhunderts. Zunächst wird die empirische Analyse von 3610 Verbclustern hinsichtlich der syntaktischen Struktur aus einer früheren Untersuchung vorgestellt. Dabei werden Aspekte wie Zeitraum, Art des Hilfsverbs, Satzkomplexität und Schreibdialekt als wichtige Faktoren bei der Verwendung unterschiedlicher Reihenfolgen angeführt. Gespiegelt an diesen Resultaten wird die Analyse mehrgliedriger hypotaktischer Verbalgruppen und parataktisch aufgebauter Verbclusters bezogen auf die Reihenfolge von finiten und infiniten verbalen Elementen vorgestellt. Zudem findet ein Vergleich mit der parataktisch aufgebauten Verbalgruppe aus der stereotypen Urkundenformel "allen denghene die dese lettren sullen sien ende horen lesen" statt. Schließlich wird der Versuch unternommen, die Verwendung der Struktur im parataktischen Verbcluster der Adressformel zu erklären.

The early 17th century comedies by Amsterdam authors Bredero, Tengnagel and Coster are written in a vernacular-like style resembling the variety spoken in Amsterdam at the time. The 17th century Amsterdam vernacular in those texts both morphologically and phonologically resembles 20th century Hollandic dialects as the Marken or Katwijk dialects or West-Frisian. The present corpus-based investigation deals with word order in verb clusters in this 17th century Amsterdam vernacular and the extent to which this deviates from present-day Standard Dutch and Hollandic dialects. The aim is twofold: first, contribute to the description of word order phenomena in older Dutch; and second, as word order is found to be an indicator for substrate influence, to investigate a Frisian substrate in Holland. Indeed correspondences between West-Frisian and Frisian word order are commonly interpreted as revealing a Frisian substrate. As far as the syntax of verb clusters is concerned, the language of the investigated comedies closely resembles present-day Dutch. While Coster seems to write rather South-Hollandish, word order in Bredero and Tengnagel resembles 20th century Katwijks and Zaans. 17th century Amsterdams does not resemble the 20th century West-Frisian which, syntactically speaking, closely resembles Frisian in being consistently verb-final, and having a distinct gerund, a restricted usage of gaan ‘go’ and blijven ‘remain’, and no ipp-effect. These findings cast doubt over the hypothesis that Frisian features used to be diffused over an area stretching much further to the south. Hence Frisian syntax does not directly testify to a historical Frisian language area in Holland, which has been reconstructed by means of other evidence.

A typical characteristic of Dutch (and other Continental West Germanic languages such as German and Frisian) is that verbs (both finite and infinite) appear together at the end of the subordinate clause in a so-called verb cluster. The ordering of the verbs within these verb clusters is not fixed in most historical stages of the Dutch language. This article studies the variable placement within verb clusters consisting of two verbs using a text corpus spanning most of the written history of the Dutch language (i.e. from 1250 until 2000). The verb order of more than six thousand verb clusters is described in the study, taking into account factors such as period, region and type of verb cluster.

Particle verbs are particle-verb combinations of which the constituting elements may or may not appear separately. This article looks at a subgroup of particle verbs with a preposition as their first part, like opbellen [to call]. When using a particle verb in a verbal cluster at the end of a clause, a Dutch speaker is free to choose between keeping particle and verb together or separating them. These possibilities are both grammatically correct. By asking participants to repeat sentences after a short pause, which were constructed to vary the distance between particle and verb (non-separated, separated by one or two verbs), we aimed to get insight into speakers’ preferred use of particle verb constructions. Results show that there is a general preference for the non-separated form of the particle verb. In addition, we observed the tendency that the larger the verbal cluster at the end of a subordinate clause, the smaller the chance that a speaker will separate the particle verb.

This paper investigates a series of Germanic verb constructions, which appear to involve the ‘wrong’ morphology on one or more of the verbs involved: Norwegian parasitic participles, Frisian upward and downward parasitic participles, and the German Skandal construction. I provide an explicit syntactic account unifying the phenomena and deriving the differences from independent differences among the languages. These apparently odd constructions are shown to be subject to specific distributional restrictions which are fully in line with standard grammatical principles of the languages under consideration. The account is based on a top-down definition of Agree, namely the claim that an unvalued feature is valued by the closest c-commanding element with the appropriate valued feature. I demonstrate that this view allows for a uniform treatment of the morphological and syntactic properties of these constructions, which, so far, have been assumed to be unrelated. Lastly, I argue that different word orders in verb clusters can be derived either by syntactic movement (in which case locality conditions have to be obeyed and new Agree(ment) relations are formed) or by pf-linearization of sister nodes (in which case no locality effects are observed and no new Agree(ment) relations are established.