The (a)-examples in (155) show that majoratives differ from adjectives modified by tetoo in that they can also be modified by modifiers specifying the distance on an implied scale between the compared entities: beduidendsignificantly, opmerkelijkremarkably, zichtbaarvisibly. The modifiers in the (b)-examples, which seem comparable to those in the (a)-examples, are possible in both cases, albeit with different meaning contributions: in (155b) the modifiers specify the distance on the implied scale between the compared entities, while in (155b') they specify the distance on the implied scale between an implicitly assumed norm and Jan’s actual size.

Majoratives differ from adjectives modified by tetoo in that they can be modified by the accented element nogeven; cf. Section 26.1.1. This is illustrated in (156). Note in passing that the primed examples are acceptable if nog is interpreted as a time adverbial, i.e. as English still. In this interpretation, which is not at issue here, nog does not need to be accented.

A remarkable case of modification, which occurs only with majoratives, is found in (157) and (158). The string hoe langer hoe A-er forms a constituent, which is clear from the fact, illustrated in the primeless (b)-examples, that it can be placed in clause-initial position as a whole; cf. the constituency test. This is also suggested by the fact, illustrated in the primed (b)-examples, that the string cannot be split.

The internal structure of the string hoe langer hoe A-er is far from clear. Perhaps we should consider the string hoe langer hoe as a lexical unit with a more or less similar meaning to the adverbs steeds/alsmaarcontinuously; cf. the discussion of (18a). Although the second majorative (beter/brutaler) is clearly the head of the complex phrase, it is not the semantic target of the modification: instead, the construction expresses that the process of getting more A is ongoing, i.e. we are dealing with a restriction related to the aspectual nature of the eventuality. This is also clear from the fact that the progressive copula wordento be cannot be replaced by the stative copula zijnto have been. In this respect, this construction again resembles adverbial phrases like steeds/alsmaarcontinuously

The examples in (160) also show that the string hoe langer hoe beter cannot easily be used in attributive position; this is supported by the fact that our Google search (Nov 8, 2022) for the string [een hoe langer hoe beter(e)] yielded no more than two relevant cases. The (a)-examples in (161) show that the same is true for the corresponding constructions with the adverb steedscontinuously; the primed example is marked without the present participle of the copula wordento become. That we are dealing with a semantic restriction is strongly suggested by the fact that the same thing can be observed in the (b)-examples in (161), where the phrase beter en beterbetter and better expresses a similar meaning as hoe langer hoe beter and steeds beter.

The examples in (162) show that the presence of the string hoe langer hoe blocks the addition of the comparative dan/als-phrase (cf. the discussion of (18a)), and that it is not compatible with other modifiers such as veelmuch.

Finally, note that the constructions in (157) and (158) should not be mixed up with the examples in (163), which may involve a (reduced) adjunct clause, and which are peculiar in that both clauses have the finite verb in clause-final position; cf. Section V10.3.1 for more details on this construction.

The first op XP na construction involves a definite numeral (usually of low cardinality). In (165a) we give an example with a superlative used in complementive position; its counterpart with the modifier zo is unacceptable (cf. *op éen na zo goed). The (b)-examples in (165) show that the PP can occur either before or after the definite determiner when the superlative is used attributively.

Example (165a) alternates with constructions such as (166a) with de instead of het. In such cases, de is the regular definite article; we are dealing with a noun phrase in which the head noun is omitted, which is indicated by [e]. This is also indicated by the attributive –e ending on the adjective best. This means that (166a) is structurally parallel to (165b), which is supported by the fact that the head noun in (165b') can also be omitted, leading to the construction in (166b).

The PP op + numeral + na must be left-adjacent to the predicatively used superlative in (165a); the (a)-examples in (167) show that the PP cannot be moved leftwards in isolation, while the (b)-examples show that the PP cannot follow the adjective; it cannot be placed immediately to its right, nor after the verbs in clause-final position. Finally, the (c)-examples show that topicalization of the adjective is possible only if the PP is pied-piped, indicating that the two form a constituent.

The examples in (168) show that (167b'&c') improve considerably with the PP as an afterthought. In this case, it is usually preceded by an intonation break, indicated by a comma, although it should be added that this break is not always easy to detect by ear.

This observation may be relevant in light of the fact that the discussion in Paardekooper (1986:655) seems to suggest that the order in (167b) is acceptable. However, since most of the examples given there involve APs that are not followed by a clause-final verb, his acceptability judgments may actually reflect the acceptability of constructions of the kind in (168b).

Example (169a) shows that neither the pre and nor the postdeterminer PP in the (b)-examples in (165) can be topicalized in isolation. This suggests that in both cases the PP forms a constituent with the noun phrase, which is further supported by the acceptability of the two (b)-examples in (169). However, the markedness of (169b) shows that the constituency of the PP and the noun phrase in (165b) is less clear than that of the PP and the superlative in (165a).

In conclusion, it can be seen from (170) that placement of the PP to the right of the noun phrase is not possible; note, however, that (170b) improves again when the PP is presented as an afterthought (i.e. preceded by an intonation break).

The discontinuous PP op ... na can also contain a complete noun phrase. This PP is especially used when the proposition expressed by the clause involves universal (∀x) or negative existential (¬∃x) quantification. The universal quantification may also be implicit, as in example (171c) without allemaalall.

The PP op NP na can also be used in the presence of a superlative, as shown in (172). This is not really surprising, since the superlative also expresses universal quantification; cf. the semantic representation in (28b). Its counterpart with the modifier zo is unacceptable, which is to be expected given the absence of universal meaning; cf. *op Peter na zo goed).

Although at first glance (172) looks very similar to the construction in (165a), we will argue below that the most likely analysis of these constructions is one in which this PP does not form a constituent with the adjective, in contrast to the PP op + numeral + na in (165). A first indication that the PPs in (165) and (172) are different is that the PP op NP na cannot be placed after the determiner in the (a)-examples in (173). Note that we find the same facts in the (b)-examples; this supports the analysis of examples of this kind in terms of N-ellipsis (cf. the discussion of (136) and (166)).

A second reason for thinking that the PPs in (165) and (172) are different is that the (a)-examples in (174) show that the PP in (172) can be moved leftwards, leaving the AP stranded. This strongly suggests that the PP and the adjective do not form a constituent in this case, but that the PP acts as an independent adverbial modifier of the clause. The unacceptability of (174b) also follows naturally from the claim that the PP functions as an independent adverbial phrase, since complementives must be left-adjacent to the verb in clause-final position; cf. Section 28.2.2, sub I. This idea is further supported by the fact that the topicalization construction in (174c) is marginal at best, although it should be noted that movement of the adjective in isolation, as in (174c'), also seems awkward. For reasons that remain unclear, placing the PP to the right of the clause-final verbs, as in (174b'), is also impossible; as usual, an intonation break before the PP improves the result.

The PP in (172) also differs from the PP in (165b). First, the examples in (175a&b) show that topicalization of the PP op Peter na in isolation is again fully acceptable, while pied piping of the PP under topicalization of the noun phrase is excluded; topicalization of the noun phrase in isolation, as in (175c), seems to yield a marginal result unless the PP in clause-final position is preceded by an intonation break.

Placing the PP immediately to the right of the noun phrase, as in (176a), is impossible, and the same seems to hold for extraposition in (176b); as usual, the result improves when the PP is presented as an afterthought (i.e. preceded by an intonation break).

The differences between the examples in (165) and (172) discussed in this subsection show that the PP op numeral na modifies the superlative, forming a constituent with either the superlative or the noun phrase containing the superlative, while the PP op NP na is not part of the AP, but acts as an adverbial modifier of (some other element in) the clause.

The superlative can also be modified by verrewegby far, as in (177a); this option is not available for adjectives modified by zo (cf. *verreweg zo goed). The modifier verreweg specifies the distance on the implied scale between the compared entities; cf. the cases of modification of the comparative in (155). Example (177b) clearly shows that the modifier and the superlative form a constituent (cf. the constituency test).

The modifier verreweg can also be used to modify a superlative in attributive position, although some speakers seem to prefer the modifier in predeterminer position. Example (178b) shows that in both cases verreweg forms a constituent with the noun phrase.

There are no other instances of modification of the superlative, which may be due to the “absolute” nature of the superlative. Note, however, that the superlative can be emphasized by morphological means, namely by adding the prefix aller- to the superlative; cf. the primeless examples in (179). However, the form aller-A-st need not be interpreted as a superlative, but can also be used as a non-comparative adjective, denoting not the highest degree, but a very high degree. This use of the form aller-A-st is sometimes called elativus, but since this term is also used for locative case, we will use the term pseudo-superlative. Some examples of this use are given in the primed examples in (179).

The emphatic and pseudo-superlatives differ in several ways. First, the attributively used emphatic superlative in (179a) is preceded by the definite article dethe; if this article is replaced by the indefinite article eena, as in (179a'), the adjective is interpreted as a pseudo-superlative. Second, the predicatively used emphatic superlative in (179b) is preceded by the element het; if het is not present, as in (179b'), the adjective is interpreted as a pseudo-superlative. Third, the examples in (179) show that the emphatic superlative can be combined with a comparative van-phrase just like the regular superlative, whereas this is excluded in the case of the pseudo-superlative. Finally, the two forms differ in the placement of the word accent; the emphatic superlative has the word accent on the prefix (àlleraardigst), while the pseudo-superlative has the word accent on the adjective (allerààrdigst).

The superlative form best can also occur as a pseudo-superlative without the prefix aller-. For some cases, see the primeless examples in (180); again, the adjective does not express the highest degree, but a very high degree of some property, which in this case is partly contextually determined. Note that the pseudo-superlative best differs from the true superlative het best in that it cannot be used as a complementive; cf. the primed examples in (180).

Although the primeless examples (180) may sound a bit old-fashioned, the pseudo-superlative best is often used in e.g. the salutation of letters: beste Jandear Jan.