Complementive adjectives function as predicates; they are predicated of one of the nominal arguments (subject or direct object) in their clause. The examples in (7) show that in Dutch they usually immediately precede the verb(s) in clause-final position. Three typical constructions with an adjectival predicate are the copular construction, the resultative construction and the vinden-construction in (7).

The complementive adjectives in (7) are part of the VP. This is clear from the fact that the verb and the AP satisfy the constituency test, according to which the position immediately preceding the finite verb in a main clause can only be occupied by a single constituent: the primeless examples in (8) show that the AP can be pied-piped by VP-topicalization. The primed examples show that the AP can also be part of a left-dislocated VP, which must also be a constituent.

The examples in (9) show that the complementive adjectives in (7) are even a necessary part of the VP: they must be present. A related feature is that the meaning of these adjectives is an inherent part of the meaning expressed by the VP as a whole. For these reasons, we will consider them to be complements of the verb, hence their name complementive adjectives.

The examples in (10) show that the ability to occur in the predicative position immediately preceding the clause-final verbs is usually not sufficient to conclude that we are dealing with an AP: the examples in (10) show that, at least in examples like (7a&c), the AP can easily be replaced by a predicatively used noun phrase such as een aardige jongen.

Supplementive adjectives differ from complementive adjectives in that they need not be present and do not contribute to the meaning expressed by the VP; instead, supplementives add their own independent meaning to the meaning of the clause. In general, supplementives are predicated of the subject or the direct object of their clause. These two possibilities are illustrated in (11). Note that (11b) is actually ambiguous between a resultative and a supplementive readin; we have opted for the most plausible reading here.

The English paraphrases in (11) are intended to express that the meanings of the VP and the supplementive adjective are relatively independent of each other. In the examples in (11), the semantic relationship between the supplementive and the VP can be loosely defined as “simultaneity”, but Section 28.3 will show that this relationship can be more complex.

As in the case of complementive adjectives, the supplementive and the VP form a constituent. This is clear from the fact, illustrated in (12), that the supplementive can be pied-piped by VP-topicalization and can be part of a left-dislocated VP.

Still, the syntactic relation between the supplementive and the verb (phrase) is of a totally different nature than the syntactic relation between the complementive and the verb: instead of acting as a complement of the main verb, the supplementive adjective acts as an adjunct of the VP. This is especially clear for supplementives that are predicated of an (agentive) subject, like dronken in (11a); example (13) shows that in such cases the supplementive can also be expressed in an en hij doet dat ... clause, which is a sufficient test for assuming adjunct status (see Section 30.2.3 for a detailed discussion of this adverbial test).

Example (14) shows that ambiguity can occasionally arise between resultative and supplementive constructions. Chapter 28 will discuss a number of means that can help to distinguish between the two constructions.

The notion of appositive is often restricted to nominal modifiers like de voorzitter van onze vereniging in (15), but we will extend the use of this term to the postnominal APs in the (b)-examples of (16), (17) and (18) below.

In some respects, appositive adjectives resemble both supplementive and attributive adjectives. Although the linear string of words in (16a) and (16b) is identical, there are reasons to think that we are dealing with a supplementive adjective and an appositive adjective, respectively. First, the examples differ in intonational contour: appositive adjectives are often preceded and followed by brief intonation breaks, indicated by commas in (16b), while this is never the case with supplementives. Second, the two examples also have a distinct difference in meaning; the English renderings show that while the supplementive can be paraphrased by an adjunct clause, the appositive is more appropriately paraphrased by a parenthetic clause.

The examples in (17) also show that the appositive is part of the noun phrase it modifies, while the supplementive acts as an independent constituent of the clause: in (17a) two independent constituents appear in the first position of the main clause and the sentence is unacceptable as a result, whereas in (17b) only a single (complex) noun phrase precedes the finite verb and the result is fully acceptable.

Thus, example (17b) shows that appositives are similar to attributive adjectives in that they are part of a complex noun phrase. However, the primeless examples in (18) show that appositives differ from attributive adjectives in that they must occur postnominally and do not inflect. The primed examples also show that appositives can modify pronouns, whereas attributive adjectives usually cannot.

Note that the adjectival appositives in the (b)-examples all have a non-restrictive interpretation. However, they can also be used as restrictors, in which case the first intonation break is missing; cf. Section 28.4, sub II. We have ignored the restrictive cases here because they behave syntactically just like the non-restrictive cases.