The most common existential quantifiers are enkele and sommige. We have already seen that the two differ in that the former can be weak or strong, while the latter is necessarily strong. That enkele can be a weak quantifier is illustrated again in the examples in (222), where enkele is used as a modifier of a subject. Under neutral intonation, the clause preferably takes the form of an expletive er construction, as in (222a); example (222b) is usually pronounced with emphatic focus accent on the quantifier. The two examples differ in interpretation: in (222a) the subject introduces new entities into domain D, while in (222b) the subject has a partitive reading; domain D already includes a set of boys, and the sentence expresses that some of these boys are walking in the street.

That the quantifier sommige is necessarily strong is clear from the fact that the expletive er construction in (223b) is impossible. This quantifier therefore cannot be used to introduce new discourse entities, but quantifies over a pre-established set of boys in domain D.

Another existential quantifier that is quite common is watsome. This quantifier is clearly weak, as shown by the fact that only the expletive er construction is acceptable in (224); the (b)-example cannot be remedied by assigning emphatic accent to the quantifier.

A striking difference between wat and enkele/sommige is that the former can easily be used as a modifier of non-count nouns, whereas the latter usually cannot. This is shown in (225).

Haeseryn et al. (1997:370) claims that in contrastive contexts sommig(e) can be combined with substance nouns like bierbeer with a “kind” interpretation, as in (226a). Although we have indeed found a small number of such cases on the internet, they strike us as distinctly odd; we much prefer the use of the plural form sommige bieren in (226b), in which case we are clearly dealing with a count noun.

Since the ability to act as a modifier of a non-count noun is also a property of degree modifiers such as veelmany/much discussed in Section 20.2.4, it might be the case that wat in (225) functions not as an existential quantifier but as a degree quantifier. Such a view could be supported by the fact that, unlike enkele and sommige, wat can be modified by degree modifiers like nogal, tamelijk, heel, aardig. This is illustrated in (227), where the cardinality of the set denoted by boeken is indeed compared to an implicit norm.

Besides the existential quantifiers discussed above, Dutch has many other formatives that can be used in a similar way. We will briefly discuss some of these formatives, starting with some simple forms and ending with some phrase-like forms.

This subsection concludes the discussion of existential quantifiers in their use as modifiers with two special uses of sommige and enkele.

The examples in (243) show that, if the context provides sufficient information, it is possible to use sommige(n) as a pronominal quantifier instead of the full quantified noun phrases sommige studenten/boekensome students/books. If the quantifier ends in a schwa, Dutch orthography requires a (mute) suffix -n on the quantifier if the elided noun is [+human]; if the elided noun is [-human], this -n is not used.

The quantifier sommige(n) is strong when used as an independent argument. It is not immediately clear whether weak quantifiers such as enkelesome can also be used as independent arguments. Consider the examples in (244). The fact that we are dealing with expletive er constructions guarantees that the quantifiers in these examples are weak. The second occurrence of er in the primed examples is so-called quantitative er, which is associated with an interpretive gap in the noun phrase, which therefore has the form [QP enkele [NP e ]]. The doubly-primed examples show that quantitative er cannot easily be omitted; this suggests that the weak quantifier enkele behaves like the cardinal numerals in that it can only act as a modifier of a noun phrase (which happens to be phonetically empty in the singly-primed examples), not as an independent argument as in the doubly-primed examples.

Note in passing that the non-main clause counterparts of the doubly-primed examples in (244) given in (245) are perfectly acceptable; the reason for this is that er can express more than one function in its location in the middle field of the clause. In the examples in (245), the single occurrence of er involves the conflation of the expletive and quantitative functions of er; cf. P37.5.3 for a detailed discussion. The acceptability of the examples in (245) is therefore not relevant to the present issue.

When enkele is used as a strong quantifier, similar complications do not arise; the primed examples in (246) are perfectly acceptable, as are those in (243) with the strong quantifier sommige.

Finally, in (247) we see that it is also possible to have quantitative er after the finite verb (in which case the spelling of enkelen in (246a') changes to enkele); this occurrence of er is like the one in (244) in that it simultaneously performs the function of expletive and quantitative er; the quantifier again functions as a weak quantifier modifying a phonetically empty noun phrase.

We conclude that existential quantifiers can be used as independent arguments only when they are strong (i.e. have a partitive meaning); they cannot be used as independent arguments when they are weak.

The evidence given in (244) for the claim that weak quantifiers such as enkelesome cannot be used as independent arguments is not conclusive, since dropping quantitative er does not lead to fully unacceptable results. Stronger support for this claim is provided by the existential quantifier wat, which cannot be used as a strong quantifier. As shown in (248), dropping quantitative er in the doubly-primed examples leads to completely unacceptable results. Note that their embedded counterparts are again acceptable (e.g. dat er wat werden verkochtthat some of them were sold), due to the fact that er performs the expletive and quantitative functions simultaneously.

Note also that the string without quantitative er in (248b'') becomes acceptable when the verb is singular, as in (249a). In this case we are not dealing with the use of the modifier wat as an independent argument, but with the colloquial form of the [-human] quantified personal pronoun ietssomething. In this use, wat can be modified by the degree modifiers heelvery and nogalquite, in which case it receives the interpretation “a lot”. Other modifiers that can be used are flinkquite and behoorlijk quite.

For the remaining simple quantifiers discussed in Subsection I, we also generally find a contrast between weak and strong quantifiers. The weak quantifiers in (250), for example, occur only when quantitative er is present, which shows that they cannot be used as independent arguments.

The forms allerlei and allerhande in (251), on the other hand, can be used as independent arguments in formal language. The independent use of these forms requires singular agreement on the finite verb: in example (251b) with quantitative er, the verb exhibits plural agreement; in example (251b') without quantitative er, the verb exhibits singular agreement. Independent allerlei and allerhande therefore pattern with wat in (249) rather than with sommige in (243).

Example (252) shows that the (formal) strong quantifier menig cannot be used as an independent argument, which may be related to the fact that the special form menigeen is used when the referent is [+human]; a corresponding [-human] form does not exist, though.

The examples in (253) show that the phrase-like quantifiers deze of gene and een of ander can also be used independently. The latter is special, however, since it can be preceded by a definite determiner and must therefore be analyzed as part (perhaps the head) of an NP. The construction as a whole is also special, since the article het does not make the noun phrase definite, which is clear from the fact that it occurs as the subject of an expletive er construction in (253b).

In this context it should be noted that het een of ander seems to be in a paradigm with het een en ander, despite the fact that een en ander cannot be used as a prenominal quantifier; cf. (254). The form het een en ander is further special in that it triggers singular agreement on the verb, despite the fact that it is semantically plural in the sense that it refers to a non-singleton set of eventualities.

Finally, example (255) shows that het nodige can also be used independently.