... construction.¹

See [4] and [13] for an account in terms of Lexical Functional grammar, [16] for an account in terms of Tree Adjoining Grammar, and [10], [28], [19], and [9] for categorial accounts.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... schemata.²

In [27], the analysis of [8] is used to account for Dutch cross-serial dependencies. The feature NPCOMP is replaced by LEX, whose distribution is again governed by the same rule-specific stipulations.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... H.³

We disallow the use of this rule if both L and R are both instantiated as the empty list. Thus, there is no place for unary rules in our grammar.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... examples:⁴

Note that we will not discuss the Dutch reflexive zich which has some peculiar properties. The current analysis ([3,15]) of this reflexive assumes that it needs to be bound by the subject of the minimal finite domain. Given that verb-clusters at best introduce non-finite domains the distribution of zich does not shed any new light on the structure of verb-cluster constructions.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

....⁵

For the moment we will assume that ``dominates'' is defined in terms of the ARG-S feature rather than the DTRS feature.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...aux2c).⁶

Some speakers accept the word order in (46c) . We come back to this issue at the end of section 6.2.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...aux2c).⁷

A way out for the binary-branching analysis would be the ensure somehow that combinations of a ``flipped'' participle and a verbal head are not marked as verb-clusters (see [1]).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... ``incorporation'':⁸

For a different analysis of separable prefixes refer to [2]

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... apply.⁹

Let us take a moment to summarize the rules and rule schemata used in this paper:

The combination-schema is defined in (8) and (9). The head-complement-schema is defined in (12). Rule 1 is an instantiation of that schema, with the ordering constraint (15) added.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.