for É Kiss, Katalin & Henk van Riemsdijk, eds. Clustering Theories1

This paper aims only to make a few brief remarks about what seem to me to be the more salient aspects of the verb cluster puzzles. I will comment briefly and somewhat selectively on points where the verb cluster data appears to bear on differences, and hence choices, among current syntactic theories. However, most questions will be left open, and I make no attempt at exhaustivity. For a far longer and more general overview of the phenomenon within Germanic see Wurmbrand (forthcoming) in addition to the introduction to this volume. 1. Verb Clusters Observationally, Dutch and German are verb-final languages. Verb-second (V2) and other movements (such as extraposition) can mask this, but when these are controlled for, the verb occurs at the right periphery of its clause, following its complement(s). In the case of verbs selecting VP or clausal complements, the net effect is that the verbs “stack up ” at the right, a fact which Mark Twain (1880) commented upon with great amusement. This is illustrated for German in (1), where the subscript numbers on the verbs illustrate their relative (semantic) hierarchical order (1 is the highest), and the verbs occur in the mirror-image order of their English counterparts in the paraphrase