/ instituut voor de Nederlandse taal / Taalportaal
bibliography glossary about
  • Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
    3.5.Special cases
    quickinfo

    This section discusses several sorts of special and/or more or less idiomatic cases of modification of the adjective. We start in Subsection I by discussing two types of postadjectival van-PPs. This is followed in II by a discussion of so-called transparent free relative clauses of the type found in Hij is wat je corpulent noemt'He is what one calls corpulent', subsections III and IV conclude by paying some attention to, respectively, the use of VP adverbs as modifiers of adjectival phrases and the modification of the adjective vol by a noun.

    readmore
    [+]  I.  Postadjectival van-PP

    In some cases, adjectives seem to be modified by a postadjectival van-PP. There are at least two types, illustrated in (328). The van-PP in (328a) expresses a restriction on the adjective “big as far as stature is concerned", and the van-PP in (328b) refers to the cause of the occurrence of the property denoted by the adjective “red with excitement". We will discuss the two cases in separate subsections.

    328
    a. groot van gestalte
      big  in  stature
    b. rood van opwinding
      red of  excitement
    [+]  A.  The construction groot van gestalte'big in stature'

    Sequences of the form A + van + noun are mostly more or less fixed collocations. The noun in the van-PP is never preceded by a determiner and cannot be modified by an adjective. If the adjective is gradable, it can be modified by an intensifier but not by an approximative/absolute adverb like vrijwel'almost' or helemaal'completely', so that we may conclude that the sequences in (329) are scalar.

    329
    a. (erg/*helemaal) groot/klein van gestalte
      very/completely  big/small  of stature
      'very big/small in stature'
    b. (zeer/*helemaal) knap van uiterlijk
      very/completely  pretty  of appearance
      'very pretty as far as the face is concerned'
    c. (vrij/*helemaal) lang/kort van stof
      rather/completely  long/brief  of subject.matter
      'rather long-winded/brief'
    d. (nogal/*helemaal) traag/snel van begrip
      rather/completely  slow/quick  of understanding
      'rather stupid/quick-witted'
    e. (vrij/*helemaal) kort van memorie
      rather/completely  short  of memory
      '(have) a rather short memory'
    f. (erg/*helemaal) trots van aard
      very/completely  pride  of nature
      '(have) a rather proud nature'

    The examples in (330) show that the A + van + N sequence can be used in the copular construction and the vinden-construction. However, the sequence cannot readily be used in resultative constructions such as (330c), which suggests that the complex APs function as individual-level predicates; cf. Section 6.2.1, sub II, ex. (25). The fact, illustrated in (330d), that the sequence cannot be used as a supplementive either is consistent with this conclusion; cf. Section 6.3, sub IV. In this respect the examples in (329) differ from sequences such as rood van opwinding'red with excitement' to be discussed in the Subsection B below, which typically function as stage-level predicates.

    330
    a. Jan is groot van gestalte.
    copular construction
      Jan  is big  in stature
    b. Ik vind Jan traag van begrip.
    vinden-construction
      I  consider  Jan  slow of understanding
    c. * De visagist maakt Jan knap van uiterlijk.
    resultative construction
      the cosmetician  makes  Jan good-looking
    d. * Knap van uiterlijk kwam Jan de kamer binnen.
    supplementive use
      well-looking  came  Jan the room  into
      'Jan entered the room well-looking.'

    An exception to the two generalizations above is the sequence gelijk van lengte in (329g): it can be modified by vrijwel'almost' or helemaal'completely' but not by the intensifier zeer and it can function as a stage-level predicate as is clear from the fact that it can be used in resultative constructions. The exceptional behavior of this sequence may be related to the fact that it is the only case that alternates with the sequence with the preposition in: gelijk in lengte'equal in length'. For this reason, we will put this exceptional case aside in the remainder of the discussion.

    331
    a. De touwtjes zijn (vrijwel/helemaal/*zeer) gelijk van lengte.
      the strings  are  almost/completely/very  equal  of  length
    b. Jan maakte de touwtjes gelijk van lengte.
      Jan made  the strings  equal  of  length

          The sequences in (329) form a constituent, which is clear from the fact that the full sequence can be placed in clause-initial position; cf. the constituency test. This is shown in (332).

    332
    a. Groot van gestalte is Jan niet.
      big of stature  is Jan not
    b. Traag van begrip vind ik Jan niet.
      slow of understanding  consider  I  Jan not

    The sequence cannot readily be split. The examples in (333) show that wh-movement or topicalization of the adjective leads to a marked result if the van-PP is stranded; the topicalization examples improve somewhat if the adjective is assigned heavy accent.

    333
    a. ? Hoe groot is Jan van gestalte?
      how big  is Jan in stature
    a'. ? Hoe traag is Jan van begrip?
      how slow  is Jan of understanding
    b. ?? Groot is Jan niet van gestalte.
      big  is Jan not  in stature
    b'. ?? Traag vind ik Jan niet van begrip.
      slow  consider  I  Jan not  of understanding

    The examples in (334) show that movement of the van-PP cannot strand the adjective either; the (a)-examples involve PP-over-V, and the (b)- and (c)-examples involve, respectively, scrambling and topicalization.

    334
    a. dat Jan groot <van gestalte> is <*van gestalte>.
      that  Jan big    in stature  is
    a'. dat ik Jan traag <van begrip> vind <*?van begrip>.
      that  I  Jan slow    of understanding  consider
    b. * Jan is van gestalte groot.
    b'. * Ik vind Jan van begrip traag.
    c. * Van gestalte is Jan niet groot.
    c'. * Van begrip vind ik Jan niet traag.

          Attributive use of the sequence A + van + N is impossible. Given the fact illustrated above that the van-PP must be right-adjacent to the adjective, this is correctly excluded by the Head-final Filter on attributive adjectives, which requires that the adjective be immediately adjacent to the modified noun; cf. Section 5.3, sub IB.

    335
    a. * een <van gestalte> grote <van gestalte> jongen <van gestalte>
      an   in stature  big  boy
    b. * een <van begrip> trage <van begrip> jongen <van begrip>
      an   of understanding  slow  boy

          Generally, the sequence A + van + N denotes a property of a human being. Consequently, it cannot occur in the partitive genitive construction iets leuks'something nice' as this construction can only denote -human entities: *iemand leuks'lit: someone nice'; cf. Section 7.2.3. Whether this fully accounts for the impossibility of the sequence in the partitive genitive construction is not clear, however, as the A + van + N sequence in (336b), which is exceptionally predicated of an -human noun phrase, cannot enter the partitive genitive construction either.

    336
    a. Jan/*de tafel is groot van stuk.
      Jan/the table  is big  of piece
      'Jan is big in bulk.'
    a'. * iets groots van stuk
    b. Het boek is knap van opzet.
      the book  is ingenious  of design
      'The book is ingeniously designed.'
    b'. * iets knaps van opzet
    [+]  B.  The construction rood van opwinding'red with excitement'

    In constructions such as (337), the van-PP does not express a restriction on the adjective but instead indicates the cause of the property denoted by the adjective; this cause is generally a mental state of the argument the adjective is predicated of, or something external that may affect the physical state of the argument the adjective is predicated of. The examples in (337) resemble those discussed in Subsection A in that they denote properties of human beings and often have an idiomatic flavor. They differ, however, in that the noun must be preceded by a definite determiner if it denotes an external cause, as in (337c); if the noun denotes a mental state, the determiner is normally absent in this type as well. It should be noted, however, that in all these types, cases in which the article is present can be found on the internet.

    337
    a. rood van (??de) opwinding
      red  of     the  excitement
      'red with excitement'
    c. blauw van *(de) kou
      blue  of    the  cold
      'blue with cold'
    b. groen van (*de) nijd
      green of     the  envy
    d. gek van (??de) angst
      mad  of     the  fear

    Modification of the adjective by means of an intensifier generally gives rise to a degraded result, whereas modification by means of the absolute modifier helemaal'completely' is easily possible. This suggests that the sequence under discussion is not gradable and perhaps is even absolute. Perhaps this is related to the fact that the sequences are metaphorical in nature; someone who is gek van angst'crazy with fear' need not be crazy. This may also account for the fact that the use of an intensifier yields a much better result in (338a) than in the other cases; one may literally turn red as the result of excitement, but one cannot become literally green as the result of envy.

    338
    a. (helemaal/?erg) rood van opwinding
      completely/very  red  of  excitement
      'completely red with excitement'
    b. (helemaal/*erg) groen van nijd
      completely/very  green of  envy
      'green with envy'
    c. (helemaal/*erg) blauw van de kou
      completely/very  blue  of the cold
      'completely blue with cold'
    d. (helemaal/*erg) gek van angst
      completely/very  mad  of  fear
      'completely crazy with fear'

    For completeness’ sake, the examples in (339) show that comparative formation gives rise to similar judgments as modification by an intensifier like erg'very', shown above.

    339
    a. Jan wordt steeds roder van opwinding.
      Jan  gets  continuously  redder  of excitement
      'Jan is continuously getting redder with excitement.'
    b'. * Jan wordt steeds groener van nijd.
    c'. * Jan wordt steeds blauwer van de kou.
    d'. * Jan wordt steeds gekker van angst.

          The sequence A + van + N(P) under discussion can be used in copular, resultative and supplementive constructions. The fact that the sequence can be used in the latter two environments is consistent with the fact that the adjective denotes a transitory property; in contrast to the constructions discussed in Subsection A, they are stage-level predicates. For some unclear reason, the sequence cannot be used in the vinden-construction.

    340
    a. Jan is rood van opwinding.
    copular construction
      Jan is red  with excitement
    b. * Ik vind Jan blauw van de kou.
    vinden-construction
      I  consider  Jan blue  of the cold
    c. Die film maakte Jan gek van angst.
    resultative construction
      that movie  made  Jan mad with fear
    d. Gek van angst rende Jan de bioscoop uit.
    supplementive use
      mad  with fear ran  Jan the cinema  out
      'Mad with fear Jan ran out of the cinema.'

          That the sequences in (337) form a constituent is clear from the examples in (341), which show that the full sequence can be placed in clause-initial position; the constituency test.

    341
    a. Rood van opwinding is Jan.
      red with excitement  is Jan
    b. Gek van angst maakte die film Jan.
      mad with fear made  that movie  Jan

    As in the constructions discussed in Subsection I, wh-movement and topicalization of the adjective here yields a degraded result if the van-PP is stranded. Note that the unacceptability of (342b) may also be due to the fact that the AP gek van angst is not gradable.

    342
    a. ?? Hoe rood is Jan van opwinding?
      how red  is Jan with excitement
    a'. ?? Rood is Jan van opwinding.
    b. * Hoe Gek is Jan van angst?
    b'. ?? Gek is Jan van angst.

    The (a)-examples in (343) show, however, that PP-over-V does seem to lead to an acceptable result, although speakers of Dutch have varying preferences with respect to placement of the PP; according to some, preverbal placement of the van-PP is preferred, whereas others strongly prefer postverbal placement (even to the point that they claim that preverbal placement is unacceptable). The (b)- and (c)-examples show that leftward movement of the PP gives rise to an ungrammatical result.

    343
    a. dat Jan rood <van opwinding> is <van opwinding>.
      that  Jan red   with excitement  is
    a'. dat die film Jan gek <van angst> maakt <van angst>
      that  that movie  Jan mad     with fear  made
    b. * Jan is van opwinding rood.
    b'. * Die film maakte Jan van angst gek.
    c. * Van opwinding is Jan rood.
    c'. * Van angst maakte die film Jan gek.

          Given that PP-over-V is possible, we may expect it to be possible for the A + van + N sequence to be used attributively with the van-PP in postnominal position, but (344) shows that this expectation is not borne out.

    344
    a. * een rode jongen van opwinding
      a red boy  of excitement
    b. * een gekke jongen van angst
      a  mad boy  of fear

    We must observe, however, that the constructions in (345a&b) are acceptable. The ungrammaticality of (345c) suggests that examples like these must be interpreted literally.

    345
    a. een rood hoofd van (de) opwinding
      a red head  of excitement
    b. blauwe handen van de kou
      blue hands  of the cold
    c. * een groene kop van nijd
      a  green head  of envy

    Possibly, the examples in (345a&b) may involve a third type of construction since the AP is not predicated of a human being but of a body part. This suggestion seems to be supported by at least two facts: first, example (345a) shows that the noun opwinding can at least optionally be preceded by a definite article and, second, the examples in (346) show that the van-PP can undergo topicalization.

    346
    a. Van opwinding werd zijn hoofd helemaal rood.
      of excitement  became  his head  completely  red
    b'. Van de kou werden zijn handen helemaal blauw.
      of the cold  became  his hands  totally  blue
    [+]  II.  Transparent free relative clauses

    Example (347) shows that adjectives can be the antecedents of non-restrictive relative clauses, in which case the relative pronoun is wat.

    347
    Jan is zeer goed in wiskunde [wati jij wel nooit ti zal worden]
      Jan is very good at math  which  you  prt  never  will  be
    'Jan is very good at math which youʼll probably never be.'

    The constructions in (348), which have been called transparent free relative clauses, are somewhat special; cf. Van Riemsdijk (2000/2006). The free relatives, which are given in square brackets, function in the same way as the adjectives corpulent and aardig'nice' would do; it is expressed that the subject is corpulent/kind, with the modification that the appropriateness of the term is open to debate.

    348
    a. Hij is [wati je ti corpulent zou kunnen noemen].
      he  is  what  one  corpulent  would  can  call
      'Heʼs what one could call corpulent.'
    b. Hij is nu [wati ik ti aardig noem].
      he  is  prt  what  I  nice  call
      'Heʼs what I call kind.'

    Like the relative clause in (347), the transparent free relative is introduced by the relative pronoun wat, which can probably be considered the logical subject of the adjective. Wat does not, however, refer to an entity in the discourse; it is clearly related to the deictic pronoun dat'that' in examples such as (349).

    349
    a. Je zou dat corpulent kunnen noemen.
      one  would  that corpulent  can  call
      'One could call that corpulent.'
    b. Ik noem dat aardig.
      I  call  that  nice
      'I call that kind.'

    The constructions in (349) are used to define or clarify the notions corpulent and aardig, which suggests that the adjectives function as second-order predicates. This seems to be supported by the fact that the relative pronoun wat in (348) clearly cannot be construed as coreferential with the subject of the main clause since it is never used to refer to human entities.
          The primeless examples in (350) show that transparent free relatives can also occur with other predicatively used categories; in these examples the predicative element is the full noun phrase een corpulente/aardige man. This is consistent with the fact that the -human pronoun dat can be used in the primed examples.

    350
    a. Hij is [wati je ti een corpulente man zou kunnen noemen].
      he is what  one  a corpulent man  would  can  call
      'Heʼs what one could call a corpulent man.'
    a'. Je zou dat een corpulente man kunnen noemen.
      one  would  that  a corpulent man  can  call
      'One could call that a corpulent man.'
    b. Hij is [wati ik ti een aardige man noem].
      he  is what  I  a nice man  call
    b'. Ik noem dat een aardige man.
      I  call  that  a kind man

          The acceptability of the attributive examples in (351) is mysterious for various reasons. In contrast to (348) and (350), the verb noemen is not preceded by a predicatively used constituent. Since this verb requires such a predicative complement, we should assume that either the adjective corpulente or the nominal projection corpulente man are acting as such, in violation of the requirement that a predicative complement precede the verbs in clause-final position; see (74) in Section 6.2.2. Further, if we assume that corpulent is the predicate of the clause, we have no account for the attributive -e ending, since predicatively used adjectives are normally not inflected; if we assume that corpulente man is the predicate of the clause, we should conclude that the determiner een can precede free relatives, which is not attested in other cases. In fact, the primed and doubly-primed examples show that free relatives of the type in (348) and (350) are both excluded after the determiner een.

    351
    a. Hij is een wat je zou kunnen noemen corpulente man.
      he  is a  what  one  would  can  call  corpulent  man
      'Heʼs a what one could call corpulent man.'
    a'. * Hij is een wat je corpulent zou kunnen noemen man.
    a''. * Hij is een wat je een corpulente man zou kunnen noemen.
    b. Hij is een wat ik noem aardige man.
      he  is a  what  I  call  nice  man
    b'. * Hij is een wat ik aardig noem man.
    b''. * Hij is een wat ik een aardige man noem.

    For more discussion and a possible solution for these mysteries, we refer the reader to Van Riemsdijk (2000/2006), who suggests that the adjective simultaneously functions as the predicate of the free relative and as the attributive modifier of the noun in the noun phrase.

    [+]  III.  VP adverbs

    Consider the examples in (352) and (353). Given the fact, illustrated in the primed examples, that the adjective can be topicalized in isolation, we conclude that the adverbial phrases are not modifiers of the adjective (although it seems that the modifiers in (352) can marginally be pied-piped under topicalization).

    352
    a. Jan is in alle opzichten gelukkig.
      Jan is in all respects  happy
    a'. Gelukkig is Jan in alle opzichten.
    b. Jan is in geen enkel opzicht geschikt.
      Jan is in no respect  suitable
    b'. Geschikt is Jan in geen enkel opzicht.
    c. Jan is op bijzondere wijze actief.
      Jan is in a special way  active
    c'. Actief is Jan op bijzondere wijze.
    353
    a. Jan is af en toe erg aardig.
      Jan is now and then  very nice
    a'. Erg aardig is Jan af en toe.
    b. Jan is soms/meestal/altijd erg aardig.
      Jan is sometimes/generally/always  very nice
    b'. Erg aardig is Jan soms/meestal/altijd.
    c. De zaak is tot op heden onopgelost.
      the case  is until now  unsolved
    c'. Onopgelost is de zaak tot op heden.

    Despite the fact that the adverbial phrases in (352) and (353) probably act as clausal adverbs, they can also be used as modifiers in the noun phrases in (354) and (355). Their ability to appear depends on the presence of the attributive adjective; if the adjective is dropped, the constructions are ungrammatical. It is not entirely clear how to account for this dependency relation if we are not dealing with a modification relation.

    354
    a. een in alle opzichten *(gelukkige) man
      an  in all respects     happy  man
    b. een in geen enkel opzicht *(geschikte) kandidaat
      an  in no respect     suitable  candidate
    c. een op bijzondere wijze *(actieve) jongen
      an  in a special way     active  boy
    355
    a. een af en toe *(erg aardige) man
      a now and then     very nice  man
    b. een soms/meestal/altijd *(erg aardige) man
      a  sometimes/generally/always     very nice  man
    c. een tot op heden *(onopgeloste) zaak
      an  until now    unsolved  case
    [+]  IV.  The sequence vol'full' + NP

    Another special case of modification involves the modification of the adjective vol'full' by a plural or a mass noun; cf. Paardekooper (1986:265ff.). Some examples are given in (356), which involve a predicative AP-complement. The fact that the noun phrase is optional strongly suggests that the adjective is the head of the predicate. The construction is very restricted; it occurs with the adjective vol only. That the adjective and the noun form a constituent is clear from the fact that they can be moved into clause-initial position.

    356
    a. Jan zette de vaas vol (bloemen).
      Jan put  the vase  full flowers
      'Jan filled the vase with flowers.'
    a'. Vol bloemen zette Jan de vaas.
    b. Jan giet de fles vol (water).
      Jan  pours  the bottle  full water
      'Jan fills the bottle full with water'
    b'. Vol water giet Jan de fles.

    The examples in (357) show that if vol modifies a noun like vaas'vase', it exceptionally follows it; again, the nominal modifier of vol is optional. In contrast to what is the case in (356), the examples in (357) seem also possible without vol: een vaas bloemen'a vase of flowers'; een fles water'a bottle of water'. In that case we are dealing, however, with the binominal construction discussed in Section N4.1.

    357
    a. een vaas vol (bloemen)
      a vase  full  flowers
      'a vase filled with flowers'
    b. een fles vol (water)
      a bottle  full    water
      'a bottle filled with water'

    The examples in (358) show that if vol is used prenominally, it can no longer be accompanied by a nominal modifier. In this connection, it should be noted that the acceptable primed examples involve compounds, and that the doubly-primed examples involve binominal constructions. In examples like these, the adjective vol can be replaced by any other appropriate attributive adjective.

    358
    a. een volle *(bloemen) vaas
      a full     flowers  vase
    a'. een volle/mooie bloemenvaas
      a full/beautiful  flower.vase
    a''. een volle/mooie vaas bloemen
      a full/beautiful  vase [of] flowers
    b. een volle *(water) fles
      a  full     water  bottle
    b'. een volle/dure waterfles
      a full/expensive  water.bottle
    b''. een volle/dure fles water
      a  full/expensive  bottle [of] water

    Finally it can be noted that the constructions in (356) and (357) alternate with the constructions in (359), in which the noun is part of a PP introduced by met'with'.

    359
    a. Jan zette de vaas vol met bloemen.
      Jan put  the vase  full  with flowers
    a'. een vaas vol met bloemen
      a vase  full   with flowers
    b. Jan giet de fles vol met water.
      Jan  pours  the bottle  full with water
    b'. een fles vol met water
      a bottle  full   with water
    References:
    • Paardekooper, P.C1986Beknopte ABN-syntaksisEindhovenP.C. Paardekooper
    • Riemsdijk, Henk C. van2000Binnenstebuitenbouw: enkele observaties over de transparante vrije relatiefzinGillis, Steven, Nuyts, Jan & Taeldeman, Johan (eds.)Met taal om de tuin geleid. Opstellen voor Georges de SchutterUniversitaire Instelling Antwerpen437-448
    • Riemsdijk, Henk C. van2000Binnenstebuitenbouw: enkele observaties over de transparante vrije relatiefzinGillis, Steven, Nuyts, Jan & Taeldeman, Johan (eds.)Met taal om de tuin geleid. Opstellen voor Georges de SchutterUniversitaire Instelling Antwerpen437-448
    • Riemsdijk, Henk van2006Free relativesEveraert, Martin & Riemsdijk, Henk van (eds.)The Blackwell companion to syntax2Malden, MA/OxfordBlackwell Publishing
    • Riemsdijk, Henk van2006Free relativesEveraert, Martin & Riemsdijk, Henk van (eds.)The Blackwell companion to syntax2Malden, MA/OxfordBlackwell Publishing
    About this article
    Author(s): Hans Broekhuis
    Category: Dutch Syntax
    Keywords: AdverbAttributive AdjectiveClauseFree Relative ClauseHead-final FilterNon-restrictive Relative ClausePartitive Genitive ConstructionRelative ClauseVP Adverb
    Version: 1.0, December 2015
    Version history:
    version editor(s) date remarks
    1.0 Hans Broekhuis December 2015
    report errorprintcite
    Jump to:
    X