Nouns containing the quantifiers ieder/elkevery and alleall can easily be used in virtually any common NP position; the examples in (188) show this for some argument positions of the verb and for nominal adverbial phrases. We will see that there are semantic differences which may affect their felicity in certain contexts, but their syntactic distribution is essentially the same.

The only exception seems to be complementive positions. This is illustrated here with the copular constructions in (189); while the use of a nominal predicate with alle is perfectly fine, a nominal predicate with ieder/elk is ruled out

The claim that elk cannot be used in nominal predicates is contradicted in Hoeksema (2013), who points to examples such as (190a). However, there is reason to reject the claim that elke week is a nominal predicate here and to assume that we are actually dealing with a temporal adverbial phrase, as suggested by the fact that (190a) alternates with the virtually synonymous expletive construction in (190b). That elke week is a temporal adverbial and not a nominal predicate is clear from the fact that it is optional in (190b) and that it does not occur in the complementive position immediately preceding the clause-final verb in (190b).

A conspicuous difference between the two types of modifiers in (187) is that noun phrases with ieder/elk are used with a singular head noun, while in the case of alle the head noun is plural. Correspondingly, subjects with ieder/elk trigger singular agreement on the finite verb, while noun phrases with alle trigger plural agreement. This is illustrated again by the examples in (191), which demonstrate the difference in grammatical number in yet another way: noun phrases with ieder/elk can only act as antecedents of singular pronouns such as zijnhis, while noun phrases with alle can only act as antecedents of plural pronouns such as huntheir. Note that (British) English uses their in both cases.

The examples in (192) show that the difference in number is not a purely grammatical matter, but is also reflected in semantics; verbs like zich verzamelento gather or omsingelento surround require their subject to be plural or to refer to a group of entities, and such verbs can take a noun phrase modified by alle as their subject, but not a noun phrase modified by elk/ieder.

The acceptability of the noun phrases modified by alle is due to the fact that they allow a collective reading, i.e. they can refer to the (contextually determined) set of individuals denoted by the head noun as a whole; thus the subjects in the primeless examples in (192) express that the property denoted by the VP applies to the set denoted by student/soldaat as a whole. Noun phrases modified by elk/ieder, on the other hand, have a distributive reading: a noun phrase such as iedere/elke student expresses that the property denoted by the VP applies to each entity in the set individually.

The examples in (192) above have shown that noun phrases modified by elk/ieder have a distributive reading, while noun phrases modified by alle can have a collective reading. Note, however, that alle is also compatible with a distributive reading: this is actually the preferred reading of example (193a), which makes it more or less semantically equivalent to (193b).

We see the ambiguity of noun phrases with alle between a collective and a distributive reading again in (194a). In its distributive reading, this example is semantically equivalent to (194b); the two sentences express that the activity of singing a song applies to each student individually; the meaning of these sentences can be satisfactorily represented by a universal operator: ∀x [Student(x) → Has sung (x, a song)]. The collective reading of (194a), on the other hand, expresses that the students sang a certain song as a group, a reading that is not available for (194b). For this collective reading of alle, the semantic representation with a universal operator seems inappropriate: strictly speaking, therefore, it is somewhat misleading to use the term universal quantifier in the case of this collective use of alle.

Example (195a) shows that the collective reading of alle studenten can be forced by adding a modifier of the type met elkaartogether or samentogether. That these modifiers force a collective reading is clear from the fact, illustrated in (195b), that they are incompatible with the distributive quantifiers ieder and elk.

The ambiguity of noun phrases with alle between a collective and a distributive reading is probably responsible for the fact that they can be used as nominal predicates, whereas noun phrases with (distributive) elk/ieder do not allow this use; cf. the examples in (189) from Subsection A.

Another difference between the two types of universal quantifier comes to the fore in noun phrases with an ordinal numeral. An example such as (196a) is perfectly acceptable, and expresses that the 100th, 200th, etc. visitor will receive a present. Example (196b), on the other hand, yields to a virtually uninterpretable result.

In noun phrases with a cardinal numeral, the use of the quantifiers ieder and elk leads to a somewhat marked result with a special interpretation: example (197a) seems to divide the set of visitors into groups of ten persons each. The use of the quantifier alle in example (197b), on the other hand, is perfectly acceptable in a context in which the cardinality of the set of visitors is 10; the quantifier alle then emphasizes that the property denoted by the VP een cadeautje krijgento get a present applies to all entities in this set. Section 19.2.1, sub V, has already shown that the quantifier and the numeral constitute a phrase; the primed (b)-examples show that this phrase can also be used as a complex predeterminer (cf. Section 20.1.2.2) or as a floating quantifier (cf. Subsection III).

The universal quantifiers elke/iedere and alle are not only used to quantify over sets of entities that are part of domain D, but they can also be used in generic statements, expressing a general rule that is assumed to be true in the speaker’s conception of the universe. As discussed in Section 18.1.1.5, we should distinguish the three types of generic statements in (198). Here we will discuss only the first two types.

When a generic statement contains a definite noun phrase, the generic statement applies generally to (entities belonging to) a particular species. Example (199a) refers to a certain species of birds, and it is claimed that this species has become extinct. In this case, the definite article cannot be replaced by the universal quantifiers alle and elke/iedere.

However, the universal quantifier alle would be acceptable in a situation where the noun denotes a species that can be divided into several subspecies: in such a case, alle would quantify over all subspecies; elk and ieder would still yield unacceptable results in such a case. The examples in (200) are intended to illustrate this.

When a generic statement contains an indefinite noun phrase, the generic statement generally applies to a prototypical member of the set denoted by the head noun. Example (201a) asserts that the prototypical zebra is striped. In this case, the indefinite article can easily be replaced by the universal quantifier alle: example (201b) simply asserts that the property of being striped applies to all zebras. The quantifiers ieder and elk can also be used in this context, but in this case the sentence has an emphatic flavor and is pronounced with the accent on the quantifier: example (201c) is best interpreted as saying that each and every entity that is a zebra is striped (cf. Haeseryn et al. 1997:349).

The grammatical gender feature [±neuter] may also serve to distinguish elk/ieder form alle, in that the form of elk/ieder depends on the value of this gender feature of the head noun, while the form of alle is invariant. This difference may be related to the fact that the head noun is singular in the former case, but plural in the latter: gender agreement between a modifier and a singular head noun is common, while the form of the modifier of a plural noun is insensitive to the gender of the noun; cf. Section 16.2, sub I.

The examples in (203) show that the de-nouns mensperson and persoonperson are exceptional in that they do not accept/need the inflectional -e ending; it is claimed that the use of elk/ieder mens is preferred by Dutch speakers while Flemish speakers prefer elke/iedere mens; cf. taaladvies.net/elk-of-elke-mens/. It seems that Dutch speakers allow both the bare and the inflected form with persoon. Note that the acceptability of the bare forms may be related to the fact that persoon (and possibly also mens) are grammatical nouns in the sense of Emonds (1985; 2000); cf. Section 19.1.1.6, sub IIIC2.

A final difference between elk/ieder and alle concerns non-count nouns. Since universal quantifiers typically quantify over a set of discrete entities, they are not expected to combine with non-count nouns. As shown in (204) for abstract non-count nouns, this expectation is indeed borne out for elke/ieder, but not for alle, which can be combined with such non-count nouns. It seems reasonable to relate this difference to the fact that only alle can give rise to a collective reading, which in the case of non-count nouns results in a “total-quantity” reading.

This does not mean that elk/ieder can never be combined with a non-count noun, but if it is, there will be a semantic clash between the distributive reading of elk/ieder and that of the non-count noun, resulting in a reinterpretation of the non-count noun as a count noun. For example, the noun phrase with the substance noun wijn in (205a) usually refers to a contextually determined quantity of wine. In (205b), on the other hand, elk/ieder triggers a count-noun interpretation of this noun, which now means “type of wine”.

The quantifier alle allows both the non-count and the count-noun interpretation: the noun wijn appears in the singular in the first case and in the plural in the second, as shown in the examples in (206).

The claim that elk/ieder can only be used with count nouns is contradicted in Hoeksema (2013) by pointing to example (207a); however, the presupposition that grond is a non-count noun is clearly wrong in view of (207b). The other counterexample to this claim is Jan laat elke hoop varenJan gives up all hope, which is clearly idiomatic in nature and actually more common with alle, and should therefore be seen as not belonging to the core grammar. In short, the claim remains unchallenged.