• Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
6.3.Quantitative er
quickinfo

We conclude this chapter with a discussion of so-called quantitative er, which is exemplified in the examples in (194). These examples show that quantitative er is associated with an interpretative gap [e] contained within a noun phrase. The reason why we discuss quantitative er in this chapter on numerals and quantifiers is that it normally requires that some quantificational element be present: the noun phrase in the second conjunct of example (194a), for example, contains the cardinal numeral drie'three' and the noun phrase in the second conjunct of example (194b) contains the quantifier veel'many'.

194
a. Jan heeft twee boeken en Piet heeft er [drie [e]].
  Jan has  two books  and  Piet has  er   three
b. Jan heeft weinig boeken maar Marie heeft er [veel [e]].
  Jan has  few books  but  Marie has  er   many

This section will discuss a number of properties of constructions with quantitative er. Since some of these properties have been illustrated earlier in this chapter, the discussion will sometimes be relatively brief.

readmore
[+]  I.  The antecedent of the interpretative gap [e]

An important condition on the occurrence of expletive er is that the gap in the associate noun phrase can be assigned an interpretation on the basis of information made available by the domain of discourse or the context. This is ensured in the examples in (194) by providing this information in the first conjunct of a coordinate structure, so that the interpretative gap [e] in (194) is construed as boeken'books'; but in the examples below we will simply tacitly assume that this condition is met.
      The examples in (195) show that, if the interpretative gap is interpreted on the basis of a previously mentioned noun phrase, the gap and its antecedent may exhibit a mismatch in number: in (195a) the antecedent is singular whereas the gap is plural, and in (195b) we are dealing with the reverse situation.

195
a. Jan heeft één boek en Piet heeft er [drie [e]].
  Jan has  one book  and  Piet has  er   three
b. Jan heeft twee boeken en Piet heeft er [één [e]].
  Jan has  two books  and  Piet has  er   one

The antecedent of the interpretative gap must be a count noun: example (194b) has shown that the quantifiers veel and weinig are compatible with quantitative er and example (196a) shows that these quantifiers can modify non-count nouns like wijn'wine', but nevertheless example (196b) is excluded (although it has been reported that some Flemish varieties do allow examples of this sort).

196
a. Jan heeft veel/weinig wijn.
  Jan has  much/little  wine
b. * Jan heeft veel wijn maar Piet heeft er [weinig [e]].
  Jan has  much wine  but  Piet has  er   little

      In the examples discussed so far the antecedent of the gap corresponds to the head of a noun phrase. The antecedent can, however, also be a larger nominal projection; cf. Blom (1977). Example (197a), for instance, shows that complements of nouns must be omitted in the quantitative er construction; the example is ungrammatical if the complement clause dat de maan om de aarde draait is present. This shows that the antecedent is not just the head noun bewijs but the lexical projection bewijs dat de aarde rond is. Example (197b) shows that the same thing holds for attributively used adjectives: the example is ungrammatical if the attributive modifier witte'white' is present.

197
a. Jan gaf [een bewijs dat de aarde rond is] en Piet gaf er [drie [e] (*dat de maan om de aarde draait)].
  Jan gave   a proof  that  the earth round  is  and  Piet gave  er   three    that  the moon  around the earth  circles
b. Jan heeft [twee zwarte katten] en Marie heeft er [drie (*witte) [e]].
  Jan has   two black cats  and  Marie has  er  three   white

Although attributive modifiers may not occur if quantitative er is present, the examples in (198) show that it is possible to have prepositional modifiers or relative clauses in such contexts.

198
a. Ik heb [twee poppen met blond haar] en Jan heeft er [drie [e] met donker haar].
  have   two dolls  with fair hair  and Jan has  er  three  with dark hair
b. Marie had [veel studenten die wilden meedoen], maar Els had er [veel [e] die weigerden].
  Marie had  many students  who  wanted  join.in but Els had er  many  who  refused

We may therefore conclude from the examples in (197) and (198) that the interpretative gap [e] is a nominal constituent larger than a head but smaller than a full noun phrase.

[+]  II.  The quantification element

The examples above have already shown that the quantificational element is normally a cardinal number or a quantifier like veel/weinig, but other quantificational elements may also occur with quantitative er: example (199a), for instance, provides a binominal construction with the quantificational nouns een paar'a couple' and een boel'a lot', and (199a) provides an example with the quantificational element genoeg'enough'; see Section 4.1.1.3, sub IVA, and Section 6.2.4, sub II, for more examples of this sort.

199
a. Ik heb nog een paar/boel boeken.
  have  still  a couple/lot  books
  'I still have a couple/lot of books.'
a'. Ik heb er nog een paar/boel.
  have  er  still  a couple/lot
  'I still have a couple of them.'
b. Ik heb nog genoeg boeken.
  have  still  enough  books
  'I still have enough books.'
b'. Ik heb er nog genoeg.
  have  er  still  enough
  'I still have enough of them.'

Although a quantificational element is present in the prototypical case, many (but not all) speakers also accept examples such as (200b). Observe that the gap [e] in (200b) must be interpreted as plural; a singular interpretation requires that it be preceded by the numeral één'one', as in (200b').

200
a. Ik heb nog een stoel/stoelen in de schuur staan.
  have  still  a chair  in the barn  stand
  'I still have a chair/chairs in the barn.'
b. % Ik heb er nog [e] in de schuur staan.
plural only
b'. Ik heb er nog [één [e]] in de schuur staan.
singular

      The presence of a numeral or quantifier is not sufficient to license the occurrence of quantitative er; there are several additional conditions that must be met. First, the associate noun phrase must be indefinite; (201a) shows that the definite counterpart of (194a) is unacceptable. Second, (201b) shows that strong quantifiers like alle are not compatible with quantitive er; see Section 6.2, sub II, for more examples. Weak quantifiers, on the other hand, normally give rise to a fully acceptable result; see Sections 6.2.3, sub II, and 6.2.4, sub II, for ample illustration.

201
a. * Jan heeft de twee boeken en Piet heeft er [de drie [e]].
  Jan has  the two books  and  Piet has  er   the three
b. * Jan heeft twee boeken van Gerard Reve en Piet heeft er [alle [e]].
  Jan has  two books by Gerard Reve  and  Piet has  er   all

      The examples in (202) suggest that quantitative er may also occur in tandem with the so-called wat voor construction. This is somewhat surprising, given that this construction is not a quantificational, but a type-denoting expression. The construction is somewhat special, however, in that it requires that the spurious article een be stressed, which is normally not possible in the wat voor construction.

202
a. Wat heeft Peter voor een/*één auto?
  what  has  Peter  for  car
  'What kind of car does Peter have?'
b. Wat heeft Peter er voor één/*een [e]?
  what  has  Peter  er  for  a
  'What kind does Peter have?'

Furthermore, the primed examples in (203) show that the nominal gap in the wat voor construction must be singular, whereas it can readily be plural in the other examples discussed above. Note that (203a') is fully acceptable with er interpreted as quantitative er, but only if the interpretive gap is construed as singular; cf. (202b). The string Wat koopt Els er voor? is acceptable if er is construed as part of a discontinuous pronominal PP er ...voor'for it', but this is irrelevant for our present discussion.

203
a. Wat koopt Els voor een boeken
  what  buys  Els for  books
  'What kind of books does Els buy?'
a'. # Wat koopt Els er voor één [e]?
  what  buys  Els  er  for  a
b. Wat koopt Els voor boeken
  what  buys  Els for  books
  'What kind of books does Els buy?'
b'. * Wat koopt Els er voor [e]?
  what  buys  Els er  for
[+]  III.  The syntactic nature of the interpretative gap

This subsection discusses the nature of the interpretative gap within the noun phrase associated with quantitative er. The three analyses in (204) come to mind and will be discussed in the next subsections.

204
a. The interpretative gap is the result of deletion: [... er ... [Num/Q [... N]]]
b. The interpretative gap is base-generated as a pronominal element, which must be licensed/bound by quantitive er: [... eri ... [Num/Q [proi]]]The interpretative gap is base-generated as a pronominal element, which must be licensed/bound by quantitive er: [... eri ... [Num/Q [proi]]]
c. The interpretative gap is the result of movement: [... eri ... [Num/Q [ti ]]]
[+]  A.  The interpretative gap is the result of deletion

A problem for a deletion analysis is that it does not explain why quantitative er must be present, given that example (156) in Section 6.2.4, sub I, has already shown that N-ellipsis is possible without quantitative er; see also Section A5.4. In fact, the contrast between the two examples in (205) shows that quantitative er is blocked in N-ellipsis contexts; (205b) is only acceptable if er is assigned a locative interpretation. For this reason, we can immediately dismiss this analysis.

205
a. Hij heeft [DP een blauwe [auto]] gekocht.
  he  has  a blue   car  bought
  'He has bought a blue one(s).'
b. # Hij heeft er [DP een blauwe [auto]] gekocht.

Barbiers (2009) has put forth a second argument against the deletion analysis. First observe that the examples in (206) show that elided nouns in N-ellipsis constructions maintain their gender and number features: the article het and the relative pronoun dat in (206a) show that the elided noun is neuter and singular; (206b) shows that changing the number of the elided noun (which is of course determined by the context) triggers changes in both the article and the relative pronoun.

206
a. Jan heeft [hetneuter,sg blauwe boekneuter,sg] datneuter,sg Peter wil hebben.
  Jan has   the  blue  book  which  Peter wants  have
b. Jan heeft [depl blauwe boekenneuter,pl] dieneuter,pl Peter wil hebben.
  Jan has   the  blue  books  which Peter wants  have

Barbiers claims that speakers of the northern varieties of Dutch fail to make a similar distinction in constructions with quantitative er: the (a)-examples in (207) show that such speakers can use the relative pronoun die both if the interpretive gap is interpreted as huis'house', which is neuter in Dutch, and if it is interpreted as auto'car', which is non-neuter. Note that we have placed Northern between quotation marks to indicate that some of our northern informants have the southern judgments, which suggests that the stratification of the distinction is in need of further research.

207
a. Jan heeft er [één huis] die je gezien moet hebben.
“Northern” speakers
  Jan has  er  one house  which  you seen must have
a'. Jan heeft er [één auto] die je gezien moet hebben.
b. Jan heeft er [één huis] dat je gezien moet hebben.
Southern speakers
  Jan has  er  one house  which  you seen must have
b'. Jan heeft er [één auto] die je gezien moet hebben.
[+]  B.  The interpretative gap is base-generated as a pronominal element

According to this analysis, proposed in Kester (1996), the presence of quantitative er is required to license some phonetically empty, base-generated pronominal-like element pro: eri ... [Num/Q [ proi ]]. If we assume that the licensing relation involves binding, the analysis can be used to account for at least some of the basic properties of the relation between quantitative er and its nominal associate. For example, given that binding requires that the binder c-command the pronominal element, we correctly predict that er must precede its associate noun phrase (if they are situated in the middle field of the clause).

208
a. Jan heeft eri [één [ proi ]] meegenomen.
  Jan has  er   one  with-taken
  'Jan has taken one of them with him.'
b. * Jan heeft [één [ proi ]] eri meegenomen.

Example (209a) further shows that the noun phrase can be placed in clause-initial position, and this also follows under the present analysis, given that the same thing is possible in the case of reflexive pronouns; the topicalized phrase is “reconstructed” into its original position (indicated by the trace tj) as far as its binding properties are concerned.

209
a. [Eén [ proi ]]j heeft Jan eritj meegenomen.
  one  has   Jan er  with-taken
b. [Voor zichzelf i]j heeft Jan ti een boek tj gekocht.
  for himself  has  Jan  a book  bought
  'John bought a book for himself.'

Finally, if we assume that the binding relation between quantitative er and its nominal associate is local in the same sense as the binding relation between a reflexive pronoun and its antecedent, we also correctly predict that er and its nominal associate must be part of the same clause: example (210) shows that placing er in some higher clause leads to ungrammaticality.

210
Jan vertelde <*eri> mij dat hij <eri> [één [ proi ]] meegenomen had
  Jan told     er  me  that  he   one  with-taken  had
'Jan told me that heʼd taken one of them with him.'

Despite this descriptive success, the suggested analysis has at least two flaws. First, it is not clear why the antecedent of pro must be er and cannot be some more meaningful element that could also indentify the semantic content of pro. Second, it is not clear how quantitative er itself is licensed; normally all elements in the clause are licensed by being in a selection or modification relation with some other elements in the clause, but this does not seem to hold for er, as it neither seems to be assigned a thematic role nor to have an obvious modification function.

[+]  C.  The interpretative gap is the result of movement

The two problems mentioned for the analyses discussed in the previous subsections are immediately solved in the movement analysis, as proposed by Coppen (1991) and Barbiers (2009), where quantitative er is claimed to pronominalize a certain part of the nominal structure. This means that er is base-generated as part of the noun phrase and subsequently moved into some NP-external position: eri ... [Num/Q [ ti ]]. That the movement is obligatory can be attributed to the more general properties of discourse linked pronouns: the examples in (211), for example, show that definite pronouns like ’m are obligatorily scrambled to the left of clausal adverbs like waarschijnlijk'probably'. See Section 8.1.3 for more discussion.

211
a. Jan heeft waarschijnlijk die blauwe auto/*’m gekocht.
  Jan has  probably  that blue car/him  bought
  'Jan has probably bought the blue car/it.'
b. Jan heeft die blauwe auto/’m waarschijnlijk gekocht.

The movement analysis also accounts for the fact that quantitative er normally precedes the noun phrase it is associated with: given that er is extracted from the noun phrase, the unacceptability of example (212b) is simply the result of freezing, the fact that a phrase from which some element is extracted is frozen in place.

212
a. Jan heeft eri [één [ ti ]] meegenomen.
  Jan has  er   one  with-taken
  'Jan has taken one of them with him.'
b. * Jan heeft [één [ ti ]]j eritj meegenomen.

Of course, some proviso must be made for the observation that in (209) topicalization of the remnant of the noun phrase is possible, by taking recourse to some notion of reconstruction. The clause-boundedness of the relation between quantitative er and its nominal associate, illustrated in (210), follows from the fact that scrambling is likewise clause-bound.
      The claim that quantitative er is pronominal and thus has referential properties is supported by the examples in (213). Example (213a) simply shows that quantitative er associated with a direct object can readily precede an indefinite indirect object. Example (213b) shows that this is not possible if the indirect object contains a noun phrase denoting the same entities as the quantified noun phrase; cf. Coppen (1991). This would immediately follow from binding condition C if er is co-referential with (that is, binds) the referential noun phrase embedded in the indirect object. The ungrammaticality of (213b) thus supports the claim that quantitative er is referential.

213
a. Hij vertelde eri iemand [drie [ ti ]].
  he  told  er  someone   three
  'He told three [= jokes] to someone.'
b. * Hij vertelde eri iemand die geen grapi kan waarderen [drie [ ti ]].
  he  told  er  someone  who  no joke  can  appreciate   three
  Intended reading: 'He told three jokes to someone who cannot appreciate a joke.'

      A powerful argument in favor of the movement analysis is that the restrictions on the relation between quantitative er and its nominal associate resemble those between a moved element and its trace. First, the examples in (214b&c) show that quantitative er cannot be associated with a single noun phrase in a coordinate structure, whereas (214d) shows that it can occur if it is associated with both noun phrases. Similar facts have been described for movement; cf. Coordinate Structure Constraint and across-the-board movement.

214
a. Jan heeft [[twee postzegels uit Thailand] en [drie postzegels uit China]].
  Jan has    two stamps from Thailand  and   three stamps from China
b. * Jan heeft eri [[twee postzegels uit Thailand] en [drie [ ti ] uit China]].
  Jan has  er    two stamps from Thailand  and   three  from China
c. * Jan heeft eri [[twee [ ti ] uit Thailand] en [drie postzegels uit China]].
  Jan has  er     two  from Thailand  and   three stamps from China
d. Jan heeft eri [[twee [ ti ] uit Thailand] en [drie [ ti ] uit China]].
  Jan has  er    two  from Thailand  and   three  from China

Second, the relation between quantitative er and its nominal associate seems to be sensitive to the same islands for extraction. Consider the examples in (215) and assume that R-extraction involves movement of an R-word from the complement position of the PP. These examples show that R-extraction is possible from complement-PPs but not from time adverbials.

215
a. Ik heb lang over mijn ontslag gepiekerd.
  I have long about my dismissal  worried
  'Iʼve worried long about my dismissal.'
a'. Ik heb eri lang [over [ ti ]] gepiekerd.
  I have there  long  about   worried
  'Iʼve worried long about it.'
b. Ik heb dat boek tijdens mijn vakantie gelezen.
  I have  that book  during my vacation  read
  'Iʼve read that book during my vacation.'
b'. * Ik heb dat boek erj [tijdens [ ti ]] gelezen.
  I have  that book  there  during  read
  Intended reading: 'Iʼve read that book during it.'

The examples in (216) show that we find the same thing with quantitative er; er can be associated with a noun phrase in a complement-PP but not in a time adverbial.

216
a. Ik heb lang over twee problemen gepiekerd.
  I have long about two problems  worried
  'Iʼve worried long about two problems.'
a'. Ik heb eri lang [over [twee [ ti ]]] gepiekerd.
  I have er long  about    two  worried
b. Ik heb dat boek tijdens twee vergaderingen gelezen.
  I have  that book  during two meetings  read
  'Iʼve read that book during two meetings.'
b'. * Ik heb dat boek eri [tijdens [twee [ ti ]]] gelezen.
  I have  that book  er  during  two  read

The examples in (217) further show that both R-extraction and the association of quantitative er require that the complement-PP precede the verb in clause-final position. Note that Ik heb er lang gepiekerd over twee is perhaps marginally acceptable with a locational reading of er, but this is of course irrelevant for our present discussion.

217
a. * Ik heb eri lang gepiekerd [over [ ti ]].
  have  there  long  worried  about
b. * Ik heb eri lang gepiekerd [over [twee [ ti ]]].
  have  er  long  worried   about   two

      The discussion above has shown that the movement analysis has much to commend itself, but it should be noted that there are also problems. The most important one is related to example (216a'), which presupposes that quantitative er can be extracted from the complement of a PP. However, there are reasons for assuming that this is normally not possible. First, consider the examples in (218), which just intend to show that R-extraction is possible from the voor-PP.

218
a. Dat brood is toch voor de lunch bedoeld.
  that bread  is prt  for lunch  intended
  'That bread is intended for lunch, isnʼt it?'
b. Dat brood is eri toch [voor [ ti ]] bedoeld.
  that bread  is there  prt   for  intended
  'That bread is intended for it, isnʼt it?'

Now consider example (219a), which differs from (218a) in that the preposition voor takes the PP-complement bij the koffie instead of the nominal phrase de lunch. Example (219b) shows that R-extraction from the voor-PP is not possible.

219
a. Die koekjes zijn toch voor bij de koffie bedoeld.
  those cookies  are  prt  for  with the coffee  intended
  'Those cookies are intended to be eaten with the coffee, arenʼt they?'
b. * Die koekjes zijn eri [voor [bij [ ti ]]] bedoeld.
  those cookies  are  there   for  with  intended

Note that R-extraction from the voor-PP gives rise to a configuration similar to the one proposed for the quantitative er construction in (217b) in the sense that a moved element is related to a trace within a complement of a preposition. The difference in acceptability of (219b) and (216a') therefore raises some doubt on the movement analysis of quantitative er, unless it can be reduced to some independent reason. One solution that comes immediately to mind is that the ungrammaticality of (219b) is due to the fact that the embedded bij-phrase is an island for extraction, but this is clearly not the case given that example (220) shows that R-extraction from the bij-phrase is possible as long as the R-word remains within the voor-PP. Since we have no further insights to offer here, we will leave this problem for future research.

220
Die koekjes zijn [voor eri [bij [ ti ]]] bedoeld.
  those cookies  are   for  there  with  intended
References:
  • Barbiers, Sjef2009Kwantitief <i>er</i> en <i>ze</i>
  • Barbiers, Sjef2009Kwantitief <i>er</i> en <i>ze</i>
  • Blom, Alied1977Het kwantitatieve <i>er</i>Spektator6387-395
  • Coppen, Peter-Arno1991Specifying the noun phraseAmsterdamThesis Publishers
  • Coppen, Peter-Arno1991Specifying the noun phraseAmsterdamThesis Publishers
  • Kester, Ellen-Petra1996The nature of adjectival inflectionUtrechtUniversity of UtrechtThesis
report errorprintcite