/ instituut voor de Nederlandse taal / Taalportaal
bibliography glossary about
  • Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
    11.3.5.Comparative (sub)deletion
    quickinfo

    This section discusses the role of wh-movement in the derivation of comparative-deletion and comparative-subdeletion constructions. The former construction is illustrated in (447), and is characterized by the fact that the comparative dan/als-phrase contains an interpretative gap, indicated by [e]. This gap receives an interpretation on the basis of (a phrase containing) an equative/comparative adjective in the matrix clause. The use of the equative form even lang'as long' in (447a) expresses that the length of table1 equals thelength of table2; the interpretative gap in the als-phrase thus receives the interpretation [Δ long] in which Δ functions as the modifier that stands for a certain degree of length. The use of the majorative form meer'more' in (447b) expresses that thenumber of books owned by Jan exceeds the number of books that Jan is able to read; the interpretative gap in the dan-phrase thus receives the interpretation [[Δ much] books] in which [Δ much] functions as a quantifier indicating quantity. Note in passing that we have placed the copular verb in the als-phrase in (447a) within parentheses to indicate that it can be (and in fact preferably is) elided under identity with the copular in the matrix clause; we will ignore this form of elision in the discussion below.

    447
    Comparative deletion
    a. Die tafel is even lang als deze tafel [e] (is).
      that table  is as long  as  this table   is
      'That table is as long as this table (is).'
    b. Jan heeft meer boeken dan hij [e] kan lezen.
      Jan has  more books  than  he  can  read
      'Jan has more books than he can read.'

    The interpretative gaps in the sentences in (447) must be syntactically present as they function as selected clausal constituents; the interpretative gap functions as a complementive in the copular construction in (447a) and as a direct object in the transitive construction in (447b). The examples in (448) show, however, that the interpretative gap can also be smaller than a clausal constituent. Example (448a) expresses that thelengthof table1 equals the width of table2, and the interpretative gap in the als-phrase thus corresponds to a subpart of the complementive; it is interpreted as the degree variable Δ of the adjectival phrase [Δ wide]. Example (448b) expresses that thenumber of books owned by Jan exceeds thenumber of CDs owned by Els, and the interpretative gap thus corresponds to a subpart of the direct object; it receives the quantifier interpretation [Δmuch] of the noun phrase [[Δ much] CDs].

    448
    Comparative subdeletion
    a. Die tafel is even lang als deze tafel [[e] breed] (is).
      that table  is  as long  as this table     wide   is
      'That table is as long as this table is wide.'
    b. Jan heeft meer boeken dan Els [[e] cd's] (heeft).
      Jan has  more books  than  Els  CDs   has
      'Jan has more books than Els has CDs.'

    This section will not provide a full discussion of comparative (sub)deletion because this is the topic of Section A4.1.3. We will focus here on the hypothesis put forward in Chomsky (1973/1977) that the interpretative gaps in the examples above are the result of wh-movement (while the wh-moved phrases themselves are subsequently deleted under "identity" with their associates in the matrix clauses). The following subsections argue that although this hypothesis seems feasible for comparative deletion, there are reasons not to accept it for comparative subdeletion. We will not discuss alternative analyses for the comparative-subdeletion construction, but refer the reader to Corver (2006) and Corver & Lechner (in prep), who discuss various proposals found in the linguistic literature.

    readmore
    [+]  I.  Comparative Deletion

    An important difference between the comparative-deletion construction and the wh-movement constructions discussed in Sections 11.3.1 to 11.3.4 is that the former does not have a phonetically visible wh-moved antecedent for the interpretative gap: if such an antecedent is present, we have to assume that it is deleted or at least left phonetically unexpressed under "identity" with (a phrase containing) the equative or comparative form that selects the dan/als-phrase. This makes it hard to firmly establish (or to refute) the claim that wh-movement is involved in the derivation of comparative-deletion constructions. In order to do this we should show that the construction exhibits at least the three characteristic properties of wh-movement listed in (449).

    449
    a. There is an obligatory interpretative gap, viz., the trace left by wh-movement.
    b. The antecedent-trace relation can be non-local in bridge contexts.
    c. The antecedent-trace relation is island-sensitive.

    That there is an interpretative gap was already shown in the introduction to this section on the basis of the meaning of the constructions in (447): we have seen, for instance, that example (450a), expresses that the number of books owned by Jan exceeds thenumber of booksthat he can read. That the gap is obligatory can furthermore be shown by the fact illustrated in (450b) that its position cannot be taken by an overt noun phrase (except, of course, for bare noun phrases with a more deeply embedded interpretative gap in comparative-subdeletion constructions).

    450
    a. Jan heeft meer boeken dan hij [e] kan lezen.
      Jan has  more books  than  he  can  read
      'Jan has more books than he can read.'
    b. * Jan heeft meer boeken dan hij de krant kan lezen.
      Jan has  more books  than  he  the newspaper  can  read

    This suggests that the comparative dan-phrase in (450a) must have a phonetically empty direct object that is associated with the overt direct object in the matrix clause containing the comparative, meer boeken'more books'. We will assume that wh-movement establishes this association by moving the (phonetically empty) phrase sufficiently close to its antecedent in the matrix clause. Because wh-movement normally results in the formation of an operator-variable chain of some sort, we will henceforth refer to the moved phrase by means of the notion empty operator (thus putting aside the question as to whether the construction involves deletion of the wh-phrase).
          There is some dispute about the precise landing site of the empty operator, which is related to the fact that the categorial status of the element dan'than' (as well as als'as') is also unclear. Although it is sometimes claimed that dan is a complementizer (that is, a subordinating conjunction), we will provisionally assume that it is a preposition-like element that is able to select a clausal complement. This seems consistent with the fact that in colloquial speech the element dan can be optionally followed by dat in examples such as (450a); because dat should clearly be analyzed as the complementizer of the embedded clause, it seems unlikely that dan has the same function. If the above is correct, we may assign example (450a) the structure in (451a). The claim that dan is preposition-like can further be supported by the fact illustrated in (451b) that it can also be followed by a noun phrase. We refer the reader to Section A4.1.3, sub III for more detailed discussion.

    451
    a. Jan heeft meer boeken [PP dan [CP Opi (dat) [TP hij ti kan lezen]]].
      Jan has  more books  than   that  he  can  read
      'Jan has more books than he can read.'
    b. Jan heeft meer boeken [PP dan [NP alleen Eline Vere van Couperus]].
      Jan has  more books  than  just  Eline Vere by Couperus
      'Jan has more books than just Eline Vere by Couperus.'

          If wh-movement is indeed involved in the derivation of comparative-deletion constructions, we expect that the interpretative gap can be embedded in complement clauses selected by bridge verbs like denken'to think' and zeggen'to say'. The examples in (452) show that this expectation is indeed borne out.

    452
    a. Jan heeft meer boeken dan ik denk dat hij [e] gelezen heeft.
      Jan has more books  than  I  think  that  he  read  has
      'Jan has more books than I think that he has read.'
    b. Jan heeft meer boeken dan ik denk dat Els zei dat hij [e] gelezen heeft.
      Jan has more books  than  I  think  that  Els said  that  he  read  has
      'Jan has more books than I think that Els said that he has read.'

    We furthermore expect comparative deletion to be excluded if the interpretative gap is embedded in an island for wh-movement. This is again borne out, as is illustrated in (453) for an interrogative and an adverbial clause, respectively. While the intended interpretations in the primed examples are perhaps hard to grasp but seem intelligible, the corresponding sentences in the primeless examples are utter gibberish.

    453
    a. * Jan heeft meer boeken dan ik vroeg [of hij [e] gelezen had].
      Jan has  more books  than  I asked   if  he  read  had
    a'. Intended reading: Jan has n books & I asked whether Jan had read m books & n > m
    b. * Jan heeft meer boeken dan ik hem bewonder [omdat hij [e] gelezen heeft].
      Jan has  more books  than  I  him  admire  because  he  read  has
    b'. Intended reading: Jan has n books & I admire Jan because he has read m books & n > m

    The island-sensitivity of comparative deletion can also be illustrated by means of the contrast between the two constructions in (454). Den Besten (1978) claims that these examples differ in that the element dan takes a clausal complement in (454a), but a nominal complement in the form of a free relative in (454b). If this proposal is on the right track, the contrast between the two examples can be attributed to the fact that wh-movement can only strand a preposition if the PP is pronominalized, that is, if it has the form waar + P.

    454
    Jan heeft meer geld verdiend ...
      Jan has  more money  earned
    a. * ... dan [CP Opi (dat) [TP zijn vrouw [PP op ti] gerekend had]].
      than   that  his wife   on  counted  had
      Intended reading: 'Jan has made more money than his wife counted on.'
    b. ... dan [NP Ø [CP waari Ø [TP zijn vrouw [PPti op] gerekend had]]].
      ... than  rel  comp  his wife  on  counted had
      'Jan has made more money than his wife had counted on.'

    The discussion above has shown that comparative deletion does indeed exhibit the three characteristic properties of wh-movement in (449): (i) the interpretative gap in the dan/als-phrase is obligatory and cannot be filled by some overt phrase (provided we put aside the comparative subdeletion-constructions); (ii) on the assumption that an empty operator is moved into the clause-initial position of the clause selected by dan, movement of this operator applies in an apparent non-local fashion in bridge contexts; (iii) movement of the empty operator is island-sensitive.

    [+]  II.  Comparative subdeletion

    Comparative-subdeletion constructions pose the same problem for establishing that wh-movement is involved in their derivation as Comparative-deletion constructions do. In order to show this, we should again prove that the construction exhibits at least the three properties of wh-movement in (449). That there is an interpretative gap was already shown in the introduction to this section on the basis of the meaning of the constructions in (448): example (455a), for instance, expresses that the number of books owned by Jan exceeds the number of CDs owned by Els. That the empty quantifier is obligatory is shown by the fact illustrated in (455b) that its position cannot be filled by an overt numeral/quantifier. This suggests that the direct object of the comparative dan-phrase in (455a) must contain a phonetically empty quantifier associated with the quantifier meer'more' of the direct object in the matrix clause meer boeken'more books'.

    455
    a. Jan heeft meer boeken dan Els [[e] cd's] (heeft).
      Jan has  more books  than  Els  CDs   has
      'Jan has more books than Els has CDs.'
    b. * Jan heeft meer boeken dan Els [duizend/veel cd's] (heeft).
      Jan has  more books  than  Els   thousand/many CDs   has

    The examples in (456) show that, as predicted by the wh-movement hypothesis, the interpretive gap can also be more deeply embedded in bridge contexts; the question mark between parentheses indicate that some speakers consider these examples slightly marked.

    456
    a. (?) Jan heeft meer boeken dan ik denk dat Els [[e] cd's] (heeft).
      Jan  has  more books  than  I think  that  Els  CDs   has
      'Jan has more books than I think Els has CDs.'
    b. (?) Jan heeft meer boeken dan ik denk dat Peter zei dat Els [[e] cd's] (heeft).
      Jan  has  more books  than I think  that Peter said  that Els  CDs   has
      'Jan has more books than I think that Peter said that Els has CDs.'

    The examples in (457) further show that comparative subdeletion is sensitive to interrogative and adjunct islands. The intended interpretations are perhaps difficult to grasp but seem intelligible, while the sentences are again utter gibberish.

    457
    a. * Jan heeft meer boeken dan ik vroeg of Els [[e] cd's] (had).
      Jan has  more books  than  I asked  if  Els  CDs   had
    a'. Intended reading: Jan has n books & I asked whether Els had m CDs & n > m
    b. * Jan heeft meer boeken dan ik Els bewonder omdat zij [[e] cd's] heeft.
      Jan has  more books  than  I  Els admire  because  she  CDs  has
    b'. Intended reading: Jan has n books & I admire Els because she has m CDs & n > m

          The data discussed so far are consistent with the wh-movement hypothesis, but there are also problems for this hypothesis. The first one is that the empty operator in (455a) is a quantifier modifying a noun phrase; the examples in (458) show that noun phrases are normally islands for wh-movement of such modifiers; movement of the quantifier hoeveel obligatorily pied pipes the containing noun phrase.

    458
    a. [Hoeveel cd's]i heeft Els ti gekocht?
      how.many CDs  has  Els  bought
      'How many CDs has Els bought?'
    b. * Hoeveeli heeft Els [ti cd's] gekocht?
      how.many  has  Els  CDs  bought

    The hypothesis that comparative-subdeletion constructions are derived by wh-movement therefore requires some special stipulation. One feasible analysis could perhaps be built in analogy to the constructions in (459), which show that quantified noun phrases like (459a) alternate with the construction in (459b) with so-called quantitative er, which replaces the lexical part of the noun phrase. We follow Coppen (1991) and Barbiers (2009) by assuming that er is extracted from the noun phrase by leftward movement (although Section N6.3 has shown that this analysis is not without problems).

    459
    a. Els heeft gisteren [veel cd's] gekocht.
      Els  has  yesterday  many CDs  bought
      'Els bought many CDs yesterday.'
    b. Els heeft eri gisteren [veel ti] gekocht.
      Els has  there  yesterday  many  bought

    The examples in (460) show that quantitative er may optionally occur in comparative-deletion constructions; cf. Bennis (1977). We can simply account for this by assuming that the interpretative gaps in the two constructions in (460) differ: the gap in (460a) receives the interpretation [[Δ much] books] while the gap in (460b) receives the interpretation [[Δ much] t], with t acting as the trace of quantitative er.

    460
    a. Jan heeft meer boeken dan hij [e] kan lezen.
      Jan has more  books  than  he  can  read
      'Jan has more books than he can read.'
    b. Jan heeft meer boeken dan hij er [e] kan lezen.
      Jan has more  books  than  he  there  can  read
      'Jan has more books that he can read.'

    This means that the two constructions in (460) are derived by comparative deletion, as the interpretative gap [e] corresponds to the full direct object in both cases. Observe that, if we follow this analysis, the noun phrase [[Δ much] t] operator must be able to be wh-moved across quantitative er; this does not pose any special problem, as is clear from the fact that it is also possible to move the remnant noun phrase in (461) across er.

    461
    [Hoeveel ti]j heeft Els eri gisteren tj gekocht?
      how.many  has  Els there  yesterday  bought
    'How many [CDs] did Els buy yesterday?'

    The acceptability of subextraction of quantitative er may lead to the conclusion that it should be possible more generally to subextract the lexical part of a noun phrase while stranding the functional part of it. We have reasons for assuming that this is possible in principle, as some varieties of (Brabantian) Dutch and German exhibit this property in so-called split-topicalization constructions such as (462b). We again assume a movement analysis (although Van Hoof, 2006, shows that this analysis is not without problems).

    462
    a. Hij heeft [NP een helehoop koeien] in de wei.
      he  has  a  lot  cows  in the field
      'He has a lot of cows in the field.'
    b. Koeieni heeft hij [NP een helehoop ti ] in de wei.
      cows  has  he  a  lot  in the field

    If lexical projections can really be extracted from their noun phrase while stranding their quantifier, this would open the possibility to reanalyze the comparative-subdeletion construction in (455a) as in (463); the underlying structure would then be approximately as in (463a), while the structure in (463b) is derived by extraction of the lexical part of the noun phrase; the surface structure in (463c) is derived by movement of the phonetically empty remnant of the noun phrase ([[Δ-much] ti]) into clause-initial position.

    463
    Jan heeft meer boeken ...
      Jan has  more books
    a. ... dan [CP (dat) [TP Els [[Δ-much] cd's] (heeft)]].
      than  that  Els  CDs   has
    b. ... dan [CP (dat) [TP Els cd'si [[Δ-much] ti] (heeft)]].
      than  that  Els CDs   has
    c. ... dan [CP [[Δ-much] ti]j (dat) [TP Els cd'sitj (heeft)]].
      than   that  Els CDs   has

    This derivation unifies comparative deletion and comparative subdeletion (for cases involving quantified noun phrases) but the cost is high; we have to make additional stipulations for Standard Dutch in order to block wh-movement of the lexical part of the noun phrase in (463b) outside the domain of comparative deletion constructions. Another reason not to follow this line of inquiry is that a wh-movement analysis of comparative subdeletion also violates other well-known restrictions on wh-movement. Example (464a) shows, for instance, that the interpretative gap can be part of a nominal complement of a PP, while the (b)-examples show that wh-movement of a subpart of a nominal complement of a PP is impossible by means of the wat voor split: wh-movement of wat obligatorily triggers pied piping of the full PP.

    464
    a. Jan kijkt naar meer tv-series dan (dat) hij naar [[e] films] kijkt.
      Jan looks  at more television.series  than   that  he  at  movies  looks
      'Jan watches more television series than he watches movies.'
    b. [Naar [wat voor films]]i kijkt Jan graag ti?
      at what for movies  looks  Jan  gladly
      'What kind of movies does Jan like to watch?'
    b'. * Wati kijkt Jan graag [naar [ti voor films]]?
      what  looks  Jan gladly   to  for films
    [+]  III.  Conclusion

    This section has looked at the role of wh-movement in comparative-deletion and comparative-subdeletion constructions. We have shown that there is good reason for assuming that comparative deletion is derived by means of wh-movement of an empty operator into the initial position of the clause selected by the prepositional-like element als/dan; this movement may be motivated by the need to place the empty operator in a sufficiently local relation with its associate, (the phrase containing) the equative/comparative adjective phrase in the matrix clause. The proper analysis of comparative subdeletion is much less clear: providing a wh-movement analysis seems to require the postulation of several ad hoc stipulations. It is therefore not surprising that this construction is still subject of ongoing debate. We refer the reader to Corver & Lechner (in prep) for a detailed discussion of the current state-of-affairs.

    References:
    • Barbiers, Sjef2009Kwantitief <i>er</i> en <i>ze</i>
    • Bennis, Hans1977Het kwantitatieve <i>er</i> in komparatief konstruktiesSpekatator6384-7
    • Besten, Hans den1978On the presence and absence of <i>wh</i>-elements in Dutch comparativesLinguistic Inquiry9641-671
    • Chomsky, Noam1973Conditions on transformationsAnderson, Stephen & Kiparsky, Paul (eds.)A festschrift for Morris HalleNew YorkHolt, Rinehart, and Winston71-132
    • Chomsky, Noam1977On <i>wh</i>-movementCulicover, Peter W., Wasow, Thomas & Akmajian, Adrian (eds.)Formal syntaxNew YorkAcademic Press71-132
    • Coppen, Peter-Arno1991Specifying the noun phraseAmsterdamThesis Publishers
    • Corver, Norbert2006Comparative deletion and subdeletionEveraert, Martin & Riemsdijk, Henk van (eds.)The Blackwell companion to syntax1Malden, MA/OxfordBlackwell Publishing
    • Corver, Norbert & Lechner, Winfriedin prepComparative deletion and subdeletionEveraert, Martin & Riemsdijk, Henk van (eds.)The Blackwell companion to syntaxMalden, MA/OxfordBlackwell Publishing
    • Corver, Norbert & Lechner, Winfriedin prepComparative deletion and subdeletionEveraert, Martin & Riemsdijk, Henk van (eds.)The Blackwell companion to syntaxMalden, MA/OxfordBlackwell Publishing
    • Hoof, Hanneke van2006Split topicalizationEveraert, Martin & Riemsdijk, Henk van (eds.)The Blackwell companion to syntax4Malden. Ma/OxfordBlackwell Publishing410-465
    About this article
    Author(s): Hans Broekhuis, Norbert Corver
    Category: Dutch Syntax
    Keywords: Bridge VerbComparative DeletionComparative SubdeletionConstructionInterrogative ClausePied PipingPreposition StrandingQuantitativeVerbWh-movement
    Version: 1.0, January 2016
    Version history:
    version editor(s) date remarks
    1.0 Hans Broekhuis, Norbert Corver January 2016
    report errorprintcite
    Jump to:
    X