Verb clustering, which was also a recurrent topic in Section 5.2 and Chapter 6, has been on the research agenda since Bech (1955) and Evers (1975), and still raises numerous questions and difficulties (both of a descriptive and of a more theoretical nature). Evers' account of verb clustering, which became the standard in early generative grammar, is as follows: in accordance with the general OV-nature of Dutch the bare infinitival clause is base-generated to the left of the matrix verb, as in (185a), and the verbal head is subsequently extracted from this clause and right-adjoined to the matrix verb, as in (185b). This head movement operation has become known as Verb Raising and is supposed to result in the formation of a monoclausal structure.

Since the early 1990's there have also been analyses that assume that infinitival clauses are base-generated to the right of the matrix verb, as in (186a). The surface order can then be derived as in (186b) by leftward movement of the non-verbal elements in the clause, as in (186b): see Coppen & Klein (1992), Den Besten & Broekhuis (1992), Zwart (1997), and many others since; Zwart (2011: Part III) provides an extensive review of proposals of this type. An alternative analysis, which is given in (186c), is based on the assumption that the German surface order bringen will is derived by leftward movement of the entire infinitival clause; see, e.g., Lattewitz (1993/1997), Broekhuis (1997a), and Barbiers (2005), The Dutch split pattern can then be derived in the same way by assuming that the infinitive has been extracted from the infinitival clause before the latter is moved leftwards (that is, by so-called remnant movement); see, e.g., Koopman & Szabolcsi (2000), Hinterhölzl (2006), and Broekhuis (2008:ch.5).

These movement approaches, of course, also suggest different solutions to the two (b)-examples in (187), in which the verb cluster is permeated by other material. Proponents of the verb movement approach, for example, may assume that not only verbal heads may right-adjoin to the matrix verb but also subparts of verbal projections (as proposed in Den Besten & Edmondson 1983), and proponents of the leftward movement approach may assume that languages vary with respect to the amount or type of leftward movement that they require.

There are also approaches to verb clustering that do not involve syntactic movement at all; see Haegeman & Van Riemsdijk (1986), Haider (2003), Kempen & Harbusch (2003), and Williams (2003). Verb clustering is probably one of the most ardently debated issues in Germanic linguistics, and the sketch given above consequently covers merely the tip of the iceberg. A good and more extensive review of the theoretical literature on verb clustering can be found in Wurmbrand (2006).