/ instituut voor de Nederlandse taal / Taalportaal
bibliography glossary about
  • Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
    2.2.5.5.The status of the postnominal PPs
    quickinfo

    Before we can apply the adjunct/complement tests provided in Section 2.2.1 to determine whether the postnominal PPs are adjuncts or complements of the picture/story noun, we have to point out that it is often not immediately clear whether a certain postnominal PP is part of the noun phrase or positioned outside the noun phrase where it functions as a complement of the verb or a restrictive adverbial phrase. Therefore, we will first consider the possible functions of these PPs in some more detail and suggest ways of distinguishing between these functions. The overall aim of this section is to find a way of establishing the status of the PPs in the various constructions, as complement of the noun or as an independent constituent (complement of the verb or restrictive adverbial phrase).

    readmore
    [+]  I.  The postnominal van-PPs of picture nouns

    Observations on PP-extraction from noun phrases headed by picture nouns are complicated by the fact that the acceptability of the resulting structures depends on the choice of the verb (Kooij & Wiers 1977/1978, Klein & Van der Toorn 1979, De Haan 1979, etc.). This is illustrated in (522) for the verbs zien'to see' and beschadigen'to damage' by means of topicalization; similar observations can be made with, e.g., relativization and questioning, but this will go unillustrated here. As shown in the primeless examples, both verbs are possible if the van-PPs follow the picture noun. If the PP is topicalized, as in the primed examples, a contrast arises between the two verbs: the result is highly questionable with the verb beschadigen'to damage', but normally fine with the verb zien'to see': an exception is (522c''), where the agentive van-PP apparently blocks topicalization of the theme.

    522
    a. Ik heb een tekening van de WestertorenTheme gezien/beschadigd.
      I  have  a drawing  of the Westertoren  seen/damaged
      'Iʼve seen/damaged a drawing of the Westertoren.'
    a'. Van de WestertorenTheme heb ik een tekening gezien/*?beschadigd.
    b. Ik heb een tekening van RembrandtAgent gezien/beschadigd.
      I  have  a drawing  of Rembrandt  seen/damaged
      'Iʼve seen/damaged a drawing by Rembrandt.'
    b'. Van RembrandtAgent heb ik een tekening gezien/??beschadigd.
    c. Ik heb een tekening van de WestertorenTh van RembrandtAg gezien/beschadigd.
      I  have  a drawing  of the Westertoren  of Rembrandt  seen/damaged
      'Iʼve seen/damaged a drawing of the Westertoren by Rembrandt.'
    c'. Van Rembrandt heb ik een tekening van de Westertoren gezien/??beschadigd.
    c''. Van de Westertoren heb ik een tekening van Rembrandt ??gezien/*?beschadigd.

    One way of accounting for these contrasts is to assume that with verbs like zien'to see' the van-PPs need not form part of the noun phrase, but may function as an independent constituent, e.g., as a complement to the verb or as a restrictive adverbial phrase. This would mean that sentences containing such verbs may be given two different structures: one in which the PP is situated within, and one in which it is situated outside the noun phrase. Of course, this difference in structure should correspond not only to a difference in syntactic behavior, but also in interpretation.
          In order to find out whether such an approach is tenable, Subsection A will consider in detail constructions of picture nouns used in combination with three types of verbs: verbs of creation like maken'to make', verbs of transfer like the ditransitive verb kopen'to buy', and verbs such as beschadigen'to damage' that denote activities that in one way or another affect their theme argument. We will look at possible structural and interpretational ambiguities of constructions involving these verbs and suggest ways to disambiguate them. Subsection B will consider the behavior of verbs like zien'to see' and vervalsen'to forge' that seem to defy classification into any of the verb types distinguished so far.

    [+]  A.  Picture nouns with maken'to make', kopen'to buy' and beschadigen'to damage'

    Picture nouns are often used in combination with verbs of creation such as schilderen'to paint' or maken'to make'. We will see that the van-PP in such constructions is ambiguous in the way indicated in (523b&b'): it can function either as a complement of the picture noun, as indicated by (523b), or as a restrictive adverbial phrase, as indicated by (523b'). We will also see that this difference in structure can be motivated by appealing to an interpretational difference; see Klein & Van der Toorn (1979).

    523
    a. Ik heb een tekening van RembrandtTheme gemaakt.
      I  have  a drawing  of Rembrandt  made
      'Iʼve made a drawing of Rembrandt.'
    b. Ik heb [NP een tekening [PP van Rembrandt]Theme] gemaakt.
    b'. Ik heb [NP een tekening] [PP van Rembrandt]Adv gemaakt.

    By way of comparison, example (524) provides similar constructions with the ditransitive verb kopen'to buy', which represents the second type of verb. Here, however, the ambiguity is of a different kind: the van-PP can be interpreted either as the agent or theme argument of the noun, as in (524b), or as a complement of the verb, as in (524b').

    524
    a. Ik heb een tekening van RembrandtAgent/Theme/Source gekocht.
      I  have  a drawing  of Rembrandt  bought
      'Iʼve bought a drawing by/of/from Rembrandt.'
    b. Ik heb [NP een tekening [PP van Rembrandt]Agent/Theme] gekocht.
    b'. Ik heb [NP een tekening] [PP van Rembrandt]Source gekocht.

    Example (525), finally, provides a construction with the affective verb beschadigen'to damage', which, we will see, only allows the interpretation with the PP as complement of the noun: the structure in (525b') does not occur.

    525
    a. Ik heb een tekening van RembrandtAgent/Theme beschadigd.
      I  have  a drawing  of Rembrandt  damaged
      'Iʼve damaged a painting by/of Rembrandt.'
    b. Ik heb [NP een tekening [PP van Rembrandt]Agent/Theme] beschadigd.
    b'. * Ik heb [NP een tekening] [PP van Rembrandt]Adv beschadigd.

    In what follows, we will refer to van-PPs that are part of the noun phrase as dependent PPs, and van-PPs that are external to the noun phrase and function as an argument of the verb or as a restrictive adverbial phrase as independent PPs. The two types of independent PPs will be distinguished by means of the labels assigned to them.

    [+]  1.  Scope of the ordinal numeral eerste'first'

    The distinction between dependent and independent van-PPs relates to a difference in interpretation, which can be made more conspicuous by adding the ordinal numeral eerste'first'; cf. also (472) and (475) in Section 2.2.5.1. The examples in (526a&b) turn out to be ambiguous: if the van-PP is dependent on the picture noun, as in the primed examples, we are not dealing with the first drawings I ever made or bought, but merely with the first drawing of Rembrandt I made/bought; if the PP is independent of the noun, as in the doubly-primed examples, the drawings in question are the first ones I made or bought. The second reading is not available for (526c), which can only mean that it is the first picture made by or depicting Rembrandt that is damaged. From this, we can conclude that the van-PP cannot be used independently with verbs like beschadigen.

    526
    a. Ik heb mijn eerste tekening van Rembrandt gemaakt.
      I  have  my first drawing  of Rembrandt  made
      'I (have) made my first drawing of Rembrandt.'
    a'. Ik heb [NP mijn eerste tekening [PP van Rembrandt]Theme] gemaakt.
    a''. Ik heb [NP mijn eerste tekening] [PP van Rembrandt]Adv gemaakt.
    b. Ik heb mijn eerste tekening van Rembrandt gekocht.
      I  have  my first drawing  by Rembrandt  bought
      'I (have) bought my first painting by/of/from Rembrandt.'
    b'. Ik heb [NP mijn eerste tekening [PP van Rembrandt]Agent/Theme] gekocht.
    b''. Ik heb [NP mijn eerste tekening] [PP van Rembrandt]Source gekocht.
    c. Ik heb mijn eerste tekening van Rembrandt beschadigd.
      I  have  my first drawing  of Rembrandt  damaged
      'I (have) damaged my first painting by/of Rembrandt.'
    c'. Ik heb [NP mijn eerste tekening [PP van Rembrandt]Agent/Theme] beschadigd.
    c''. *? Ik heb [NP mijn eerste tekening] [PP van Rembrandt]Adv beschadigd.
    [+]  2.  Negation

    Insertion of the negator niet'not' can also be used to disambiguate these sentences, provided that we assume that it can only be placed between the noun and an independent van-PP, i.e., that it cannot occur internal to the noun phrase. This correctly predicts that (527a&b) will only be acceptable on the reading that the picture involved is the first I ever made/bought, whereas the construction in (527c) is unacceptable, as neither of the two possible structures is available.

    527
    a. Ik heb mijn eerste tekening niet van Rembrandt gemaakt.
      I  have  my first drawing  not  of Rembrandt  made
      'I didnʼt make my first drawing of Rembrandt.'
    a'. * Ik heb [NP mijn eerste tekening niet van Rembrandt]Theme gemaakt.
    a''. Ik heb [NP mijn eerste tekening] niet [PP van Rembrandt]Adv gemaakt.
    b. Ik heb mijn eerste tekening niet van Rembrandt gekocht.
      I  have  my first drawing  not  of Rembrandt  bought
      'I didnʼt buy my first drawing from Rembrandt.'
    b'. * Ik heb [NP mijn eerste tekening niet van Rembrandt]Agent/Theme gekocht.
    b''. Ik heb [NP mijn eerste tekening] niet [PP van Rembrandt]Source gekocht.
    c. * Ik heb mijn eerste tekening niet van Rembrandt beschadigd.
      I  have  my first drawing  not  of Rembrandt  damaged
    c'. * Ik heb [NP mijn eerste tekening niet van Rembrandt]Agent/Theme beschadigd.
    c''. * Ik heb [NP mijn eerste tekening] niet [PP van Rembrandt]Adv beschadigd.
    [+]  3.  Topicalization

    Topicalization of the noun phrase can also be used to distinguish the two readings. Of course, since van-PP can be dependent on the noun with all verbs, it does not come as a surprise that it is always possible to topicalize the van-PP along with the noun phrase. Since in Dutch only a single constituent can be placed before the verb in sentence initial position (the constituency test), the prediction is that only the structures in the singly-primed examples are acceptable. This predication is again borne out: the sentences in the primeless examples can only mean that in 1982 it was the first time I made/bought/damaged a painting of Rembrandt; the reading corresponding to the doubly-primed examples that this picture was the first I ever made/bought/damaged is not available.

    528
    a. Mijn eerste tekening van Rembrandt heb ik gemaakt in 1982.
      my first drawing  of Rembrandt  have  I  made  in 1982
      'My first drawing of Rembrandt I made in 1982.'
    a'. [NP Mijn eerste tekening [PP van Rembrandt]Theme]i heb ik ti gemaakt in 1982.
    a''. * [NP Mijn eerste tekening]i [PP van Rembrandt]Adv-j heb ik titj gemaakt in 1982.
    b. Mijn eerste tekening van Rembrandt heb ik gekocht in 1982.
      my first drawing  of Rembrandt  have  I  bought  in 1982
      'My first drawing by/of Rembrandt I bought in 1982.'
    b'. [NP Mijn eerste tekening [PP van Rembrandt]Agent/Theme]i heb ik ti gekocht in 1982.
    b''. * [NP Mijn eerste tekening]i [PP van Rembrandt]Source-j heb ik titj gekocht in 1982.
    c. Mijn eerste tekening van Rembrandt heb ik beschadigd in 1982.
      my first drawing  of Rembrandt  have  I  damaged  in 1982
      'My first drawing by/of Rembrandt I damaged in 1982.'
    c'. [NP Mijn eerste tekening [PP van Rembrandt]Agent/Theme]i heb ik ti beschadigd in 1982.
    c''. * [NP Mijn eerste tekening]i [PP van Rembrandt]Adv-j heb ik titj beschadigd in 1982.

          Next, let us consider the possibilities for topicalizing the picture noun only, leaving the PP in the original position. Here we would expect only the doubly-primed structures to yield acceptable results. The examples in (529) show that this prediction is also borne out: the examples in (529a&b) can only express that the drawing in question is the very first one I made or bought, which means that these sentences must be assigned the structures in (529a''&b''). The fact that (529c) does not allow this type of topicalization again shows that the PP cannot be used independently with the verb beschadigen'to damage'.

    529
    a. Mijn eerste tekening heb ik van Rembrandt gemaakt.
      my first drawing  have  I  of Rembrandt  made
      'My first drawing I made of Rembrandt.'
    a'. * [NP Mijn eerste tekening]i heb ik [NPti [PP van Rembrandt]Theme] gemaakt.
    a''. [NP Mijn eerste tekening]i heb ik ti [PP van Rembrandt]Adv gemaakt.
    b. Mijn eerste tekening heb ik van Rembrandt gekocht.
      my first drawing  have  I  of Rembrandt  bought
      'My first drawing I bought from Rembrandt.'
    b'. * [NP Mijn eerste tekening]i heb ik [NPti [PP van Rembrandt]] gekocht.
    b''. [NP Mijn eerste tekening]i heb ik ti [PP van Rembrandt]Source gekocht.
    c. * Mijn eerste tekening heb ik van Rembrandt beschadigd.
      my first drawing  have  I  of Rembrandt  damaged
    c'. * [NP Mijn eerste tekening] heb ik [NPti [PP van Rembrandt]] beschadigd.
    c''. * [NP Mijn eerste tekening] heb ik ti [PP van Rembrandt]Adv beschadigd.

    Finally, let us look at the possibilities for topicalizing the van-PP. Not surprisingly, it appears that only independent van-PPs can be topicalized. The relevant examples can be found in (530). The examples in (530a&b) indeed seem to have the interpretation associated with the independent use of the PP: example (530a) expresses that the very first drawing I ever made was one of Rembrandt, and example (530b) expresses that the drawing was bought from Rembrandt (source) and that it was indeed the first I ever bought. As expected by now, example (530c) is degraded: the reading associated with (530c') is at best marginally acceptable in the (unlikely) context where I am planning to damage several of my drawings of/by Rembrandt.

    530
    a. Van Rembrandt heb ik mijn eerste tekening gemaakt.
      of Rembrandt  have  I  my first drawing  made
      'Of Rembrandt I (have) made my first drawing.'
    a'. * [PP Van Rembrandt]Theme-i heb ik [NP mijn eerste tekening ti] gemaakt.
    a''. [PP Van Rembrandt]Adv-i heb ik [NP mijn eerste tekening] ti gemaakt.
    b. Van Rembrandt*Agent/*Theme/Source heb ik mijn eerste tekening gekocht.
      of Rembrandt  have  I  my first drawing  bought
      'From Rembrandt I bought my first drawing.'
    b'. * [PP Van Rembrandt]i heb ik [NP mijn eerste tekening ti] gekocht.
    b''. [PP Van Rembrandt]Source-i heb ik [NP mijn eerste tekening] ti gekocht.
    c. ?? Van RembrandtTheme/Agent heb ik mijn eerste tekening beschadigd.
      of Rembrandt  have  I  my first drawing  damaged
      'Iʼve damaged my first drawing of Rembrandt.'
    c'. ?? [PP Van Rembrandt]Theme/Agent-i heb ik [NP mijn eerste tekening ti] beschadigd.
    c''. * [PP Van Rembrandt]Adv-i heb ik [NP mijn eerste tekening] ti beschadigd.

    It should be noted, however, that it is possible to force the reading associated with (530a'), according to which the drawing in question is not my very first, but the first I made of Rembrandt, by adding an adverb like nu'now' or inmiddelsook'by now also'. Clearly, this cannot be accounted for by assuming that only PPs functioning as independent constituents can be topicalized; we therefore assume the same thing to be possible with complements of picture nouns.

    531
    a. Van Rembrandt heb ik nu mijn eerste tekening gemaakt.
      of Rembrandt  have  I  now  my first drawing  made
      'Of Rembrandt I (have) made my first drawing now.'
    b. [PP Van Rembrandt]Theme-i heb ik nu [NP mijn eerste tekening ti] gemaakt.
    b'. * [PP Van Rembrandt]Adv-i heb ik nu [NP mijn eerste tekening] ti gemaakt.
    [+]  B.  Picture nouns with zien'to see' and vervalsen'to forge'

    This subsection discusses constructions with the non-affective verbs zien'to see' and vervalsen'to forge' to see whether there are reasons for assuming that these, too, can be combined with an independent van-PP. The answer will be negative; we will show that the structure in (532b') is excluded.

    532
    a. Ik heb een tekening van Rembrandt gezien/vervalst.
      I  have  a drawing  of Rembrandt  seen/forged
      'Iʼve seen/forged a drawing by/of Rembrandt.'
    b. Ik heb [een tekening [van Rembrandt]Agent/Theme] gezien/vervalst.
    b'. * Ik heb [een tekening] [van Rembrandt] gezien/vervalst.

    The first reason for this conclusion is that adding the ordinal numeral eerste does not create ambiguity. The most plausible interpretation of the examples in (533a), which are perhaps more easily interpretable with vervalsen than with zien, is that the drawing is the first drawing of/by Rembrandt I ever saw/forged. An alternative interpretation is that I saw or forged a certain drawing of/by Rembrandt for the first time. The difference between these two readings relates to a difference in the specificity of the referent of the phrase mijn eerste tekening van Rembrandt; in either case the drawing in question is my first drawing of Rembrandt, so that we are dealing with dependent PPs in both cases. This suggests that structure (533b') is not available.

    533
    a. Ik heb mijn eerste tekening van Rembrandt gezien/vervalst.
      I  have  my first drawing  of Rembrandt  seen/forged
      'Iʼve seen/forged my first drawing of Rembrandt.'
    b. Ik heb [mijn eerste tekening [van Rembrandt]Agent/Theme] gezien/vervalst.
    b'. * Ik heb [mijn eerste tekening] [van Rembrandt] gezien/vervalst.

    On the basis of these facts, we expect placement of the negator niet in between the picture noun and the van-PP to yield an unacceptable result. The examples in (534) show that this is indeed borne out: both sentences are semantically ill-formed, regardless of the semantic role of the constituent Rembrandt (agent, theme, possessor).

    534
    a. * Ik heb een tekening niet van Rembrandt gezien (maar van Frans Hals).
      I  have  a drawing  not of Rembrandt  seen   but of Frans Hals
    b. * Ik heb mijn eerste tekening niet van Rembrandt gezien/vervalst (maar van Frans Hals).
      I  have  my first drawing  not of Rembrandt  seen/forged  but of Frans Hals

    Topicalization, however, provides equivocal evidence. The fact illustrated by (535a) that topicalization of the entire noun phrase is possible again confirms that the structure in (533b) is available, but evidence in favor of the ungrammaticality of the structure in (533b') is less firm: the claim is supported by the fact illustrated by (535b) that topicalization of the noun phrase in isolation is impossible, but contradicted by the fact illustrated by (535c) that topicalization of the van-PP in isolation is allowed.

    535
    a. Mijn eerste tekening van Rembrandt heb ik gezien.
      my first drawing  of Rembrandt  have  I  seen
      'Iʼve seen my first drawing by/of Rembrandt.'
    b. * Mijn eerste tekening heb ik van Rembrandt gezien (niet van Frans Hals).
      my first drawing  have  I  of Rembrandt  seen   not of Frans Hals
    c. Van Rembrandt heb ik mijn eerste tekening gezien.
      of Rembrandt  have  I  my first drawing  seen

    It must further be noted that examples behave slightly differently with respect to the negation test and topicalization if we use a determiner like zoʼn'such a' or een dergelijke'a similar'. Example (536a) shows that in that case negation can be placed between the noun phrases and the van-PP, and given that we can place all sort of adverbial material (like gisteren nog'only yesterday') between the noun phrase and the PP, it is very likely that we are dealing here with an independent van-PP. This suggestion is also supported by the fact illustrated in (536b) that topicalization of the picture noun may strand the PP. It seems that this difference is again related to the interpretation of the noun phrase: the noun phrase in (534a) refers to a specific token (hence the unacceptability: one and the same drawing cannot be of/by two separate persons), whereas the noun phrases in (536) refer to a type of drawing.

    536
    a. Ik heb zoʼn/een dergelijke tekening niet van RembrandtAgent/Theme gezien.
      I have  such a/a similar  drawing  not of Rembrandt  seen
      'Such a/A similar drawing Iʼve seen not by/of Rembrandt.'
    b. Zoʼn/Een dergelijke tekening heb ik (ook) van RembrandtAgent/Theme gezien.
      such a/a similar drawing  have  I   also  of Rembrandt  seen
    [+]  C.  Conclusion

    It seems that we may tentatively conclude on the basis of the preceding discussion that there is no reason for assuming that sentences with the verbs zien'to see' or vervalsen'to forge' are structurally different from those with the verb beschadigen'to damage'. Whereas there is clear syntactic and semantic evidence for assuming that verbs like maken and kopen clearly may occur in two different syntactic frames, such evidence is lacking in the case of verbs like zien and vervalsen. The fact that the latter verbs allow topicalization of the PP therefore suggests that topicalization is not restricted to independent van-PPs, but can also apply to complements of picture nouns. The test of PP-extraction may therefore still be a valid means of distinguishing between complements and adjuncts within the noun phrase.

    [+]  II.  The postnominal van/over-PPs of story nouns

    As in the case of picture nouns, observations on PP-extraction from noun phrases headed by story nouns are complicated by the fact that the acceptability of the resulting structures depends on the choice of the verb. Constructions with story nouns used in combination with verbs of communication like schrijven'to write' or lezen'to read' will be shown to be both structurally and interpretationally ambiguous; the over-PP in (537a), for example, can be interpreted as being the complement of the story noun or as functioning as an independent constituent of the clause.

    537
    a. Jan heeft een boek over WO II geschreven.
      Jan  has  a book  about WW II  written
      'Jan has written a book about World War II.'
    b. Jan heeft [NP een boek [PP over WO II]Theme] geschreven.
    b'. Jan heeft [NP een boek] [PP over WO II]Theme geschreven.

    The same thing can be said of transfer verbs like kopen'to buy' in (524a), in which the van-PP can be interpreted either as the agent of the noun or as the source argument of the verb. The preferred reading of (538a) depends on the knowledge of the speaker: those who know that Jan Arends is a writer will probably opt for the reading in (538b) whereas those who do not will opt for the reading in (538b'). We will not discuss this any further here since this would simply repeat the discussion on picture nouns in Subsection I.

    538
    a. Ik heb een boek van Jan ArendsAgent/Source gekocht.
      I  have  a book  of Jan Arends  bought
      'Iʼve bought a book by/from Jan Arends.'
    b. Ik heb [NP een boek [PP van Jan Arends]Agent/Theme] gekocht.
    b'. Ik heb [NP een boek] [PP van Jan Arends]Source gekocht.

    Affective verbs like verbranden'to burn' normally require construal of the van/over-PP as an argument of the story noun. This means that an example such as (539a) can only be interpreted with the structure in (539b); the structure in (539b') is unacceptable.

    539
    a. Jan heeft een boek over WO II verbrand.
      Jan  has  a book  about WW II  burnt
      'Jan has burnt a book about World War II.'
    b. Jan heeft [NP een boek [PP over WO II]Theme] verbrand.
    b'. * Jan heeft [NP een boek] [PP over WO II] verbrand.

    In what follows, we will again refer to van-PPs that are part of the noun phrase as dependent PPs and to van-PPs that are external to the noun phrase and function as an argument of the verb or as a restrictive adverbial phrase as independent PPs. The overall aim of this subsection will be to find a way of establishing the status of the van- and over-PPs in the various constructions: in Subsection A we will discuss constructions in which only an over-PP is present, and in Subsection B construction in which only a van-PP is expressed; Subsection C will look at the verbs zien'to see' and vertalen'to translate' that seem to defy classification into any of the verb types distinguished so far. Finally, Subsection D will provide a number of conclusions.

    [+]  A.  Story nouns + over-PPs

    This subsection discusses the difference between sentences such as (540a) that contain a verb of communication like schrijven'to write' and sentences such as (540b) that contain an affective verb like verbranden'to burn'. We will show that the former are ambiguous between readings with a dependent and an independent PP, whereas the latter are only possible with a dependent PP.

    540
    a. Jan heeft een boek over WO II geschreven.
      Jan  has  a book  about WW II  written
      'Jan has written a book about World War II.'
    b. Jan heeft een boek over WO II verbrand.
      Jan  has  a book  about WW II  burnt
      'Jan has burnt a book about World War II.'
    [+]  1.  Scope of the ordinal numeral eerste'first'

    The distinction between dependent and independent over-PPs relates to a difference in interpretation, which can be made more prominent by adding the ordinal numeral eerste'first'. Consider the ambiguous example in (541a). If the over-PP is dependent on the story noun, as in the primed example, we are not dealing with the first book that Jan has ever written; it is merely the first book on World War II that he wrote. If the PP is independent of the noun, as in the doubly-primed example, the book in question is the very first one Jan wrote. Sentence (541b) does not allow this type of ambiguity: the noun phrase must refer to the first book on World War II that Jan wrote or acquired (depending on the interpretation of the possessive pronoun zijn as an agent or a possessor).

    541
    a. Jani heeft zijni eerste boek over WO II geschreven.
      Jan  has  his first book  about WW II  written
      'Jan has written/wrote his first book about World War II.'
    a'. Jani heeft [NP zijni-Agent eerste boek [PP over WO II]Theme] geschreven.
    a''. Jani heeft [NP zijni-Agent eerste boek] [PP over WO II]Theme geschreven.
    b. Jani heeft zijni eerste boek over WO II verbrand.
      Jan  has  his first book  about WW II  burnt
      'Jan has burnt his first book about World War II.'
    b'. Jani heeft [NP zijni-Agent/Poss eerste boek [PP over WO II]Theme] verbrand.
    b''. * Jani heeft [NP zijni-Agent/Poss eerste boek] [PP over WO II]Theme verbrand.
    [+]  2.  Negation

    Insertion of the negator niet'not' can also be used to disambiguate these sentences if we assume that negation cannot occur within the noun phrase, that is, that it can only be placed between the noun and an independent over-PP. This correctly predicts that (542a) will be acceptable on the reading that the first book that Jan ever wrote was about World War II, whereas the construction in (542b) is unacceptable, as neither of the two possible structures is available.

    542
    a. Jani heeft zijni eerste boek niet over WO II geschreven.
      Jan  has  his first book  not  about WW II  written
      'Jan didnʼt write his first book about World War II.'
    a'. * Jani heeft [NP zijni-Agent eerste boek niet over WO IITheme ] geschreven.
    a''. Jani heeft [NP zijni-Agent eerste boek] niet [PP over WO II]Theme geschreven.
    b. * Jani heeft zijni eerste boek niet over WO II verbrand.
      Jan  has  his first book  not  about WW II  burnt
    b'. * Jani heeft [NP zijni-Agent/Poss eerste boek niet over WO IITheme ] verbrand.
    b''. * Jani heeft [NP zijni-Agent/Poss eerste boek] niet [PP over WO II]Theme verbrand.
    [+]  3.  Topicalization

    Topicalization can also be used to distinguish the two readings. Of course, since over-PPs can be dependent with both verb types, it does not come as a surprise that it is always possible to topicalize both the noun phrase and the PP. Since in Dutch only a single constituent can be placed before the verb in sentence initial position (the constituency test), we predict that only the constructions in the singly-primed examples are acceptable. This prediction is borne out: sentence (543a) can only mean that it is the first time Jan has written a book about the World War II, not that it is the first book he ever wrote.

    543
    a. Zijn eerste boek over WO II heeft Jan in 1982 geschreven.
      his first book  about WW II  has  Jan  in 1982  written
      'Jan wrote his first book about World War II in 1982.'
    a'. [NP Zijn eerste boek [PP over WO II]Theme ]i heeft Jan ti in 1982 geschreven.
    a''. * [NP Zijn eerste boek]i [PP over WO IITheme ]j heeft Jan titj in 1982 geschreven.
    b. Zijn eerste boek over WO II heeft Jan in 1982 verbrand.
      his first book  about WW II  has Jan  in 1982  burnt
      'Jan burnt his first book about World War II in 1982.'
    b'. [NP Zijn eerste boek [PP over WO II]Theme]i heeft Jan ti in 1982 verbrand.
    b''. * [NP Zijn eerste boek]i [PP over WO IITheme]j heeft Jan titj in 1982 verbrand.

          Next, let us consider the possibilities for topicalizing the story noun only, leaving the PP in the original position. Here we would expect only the doubly-primed examples to yield acceptable results. The examples in (544) show that this prediction is borne out: example (544a) can only express that the book in question is the very first Jan wrote, and example (544b) does not allow this type of topicalization, as the only structure that would allow this kind of topicalization (the doubly-primed one) is not available.

    544
    a. Zijn eerste boek heeft Jan over WO II geschreven.
      his first book  has  Jan  about WW II  written
      'Jan has written his first book about World War II.'
    a'. * [NP Zijn eerste boek]i heeft Jan [NPti [PP over WO II]Theme] geschreven.
    a''. [NP Zijn eerste boek]i heeft Jan ti [PP over WO II]Theme geschreven.
    b. * Zijn eerste boek heeft Jan over WO II verbrand.
      his first book has  Jan  about WW II  burnt
    b'. * [NP Zijn eerste boek]i heeft Jan [NPti [PP over WO II]Theme] verbrand.
    b''. * [NP Zijn eerste boek]i heeft Jan ti [PP over WO II]Theme verbrand.

          Finally, let us look at the possibilities for topicalizing of the over-PP. Not surprisingly, the idea is that only independent van-PPs can be topicalized. The relevant examples can be found in (545). Example (545a) seems to have the interpretation associated with the independent use of the PP: it expresses that the very first book Jan wrote was about the World War II. As expected, example (545b) is degraded due to the fact that the over-PP can only be interpreted as a dependent of the story noun.

    545
    a. Over WO II Theme heeft Jan zijn eerste boek geschreven.
      about WW II  has  Jan his fist book  written
      'Itʼs about World War II that Jan has written/wrote his first book.'
    a'. * [PP Over WO II]i heeft Jan [NP zijn eerste boek ti] geschreven.
    a''. [PP Over WO II]i heeft Jan [NP zijn eerste boek] ti geschreven.
    b. * Over WO IITheme heeft Jan zijn eerste boek verbrand.
      about WW II has  Jan  his fist book  burnt
    b'. * [PP Over WO II]i heeft Jan [NP zijn eerste boek ti] verbrand.
    b''. * [PP Over WO II]i heeft Jan [NP zijn eerste boek] ti verbrand.

    It should be noted, however, that it is possible to force the reading associated with (545a'), according to which the book in question is not the first Jan wrote, but the first he wrote on World War II, by adding an adverb like pas onlangs'only recently'. Clearly, this cannot be accounted for by assuming that only PPs functioning as independent constituents can be topicalized, and we should therefore assume the same thing to be possible with complements of story nouns.

    546
    a. Over WO IITheme heeft Jan pas onlangs zijn eerste boek geschreven.
      about WW II  has  Jan only recently  his fist book  written
      'Itʼs about World War II that Jan has recently written/wrote his first book.'
    b. [PP Over WO II]i heeft Jan pas onlangs [NP zijn eerste boek ti] geschreven.
    b'. * [PP Over WO II]i heeft Jan [NP zijn eerste boek] ti geschreven.

          A similar effect can be achieved in the constructions in (547) involving R-extraction. Again, the assumption would be that extraction is possible from a PP complementing the VP, but not from a PP functioning as complement of the noun. The preferred interpretation of (547a) is in accordance with this assumption: the question concerns the first book that Jan ever wrote. The alternative reading becomes available, however, if we add the adverb onlangs: now the question concerns the new subject that Jan recently wrote a book about. Clearly, this cannot be accounted for by assuming that R-extraction is only possible with PPs that function as complements of the verb.

    547
    a. WaarTheme heeft Jan zijn eerste boek over geschreven?
      where  has  Jan his first book  about  written
      'What is it Jan wrote his first book about?'
    b. WaarTheme heeft Jan onlangs zijn eerste boek over geschreven?
      where  has  Jan recently  his first book  about  written
      'What is it Jan has recently written his first book about?'
    [+]  B.  Story nouns + van-PPs

    This subsection considers story noun constructions in which only the van-PP is mentioned. Example (548) gives examples involving the communication verb lezen'to read' and the affective verb verbranden'to burn'. The primed and doubly-primed examples are the structures associated with, respectively, the dependent and independent use of the van-PP. In what follows we will see that (548b) with the affective verb verbranden'to burn' allows only a dependent interpretation of the PP in (548b'); (548b'') does not occur.

    548
    a. Jan heeft een boek van Huizinga gelezen.
      Jan has  a book  of Huizinga  read
      'Jan has read a book by Huizinga.'
    a'. Jan heeft [NP een boek [PP van Huizinga]Agent] gelezen.
    a''. Jan heeft [NP een boek] [PP van Huizinga]Adv gelezen.
    b. Jan heeft een boek van Huizinga verbrand.
      Jan has  a book  of Huizinga  burnt
      'Jan has burnt a book by Huizinga.'
    b'. Jan heeft [NP een boek [PP van Huizinga]Agent] verbrand.
    b''. * Jan heeft [NP een boek] [PP van Huizinga] verbrand.
    [+]  1.  Scope of the ordinal numeral eerste'first'

    The distinction between dependent and independent van-PPs relates to a difference in interpretation, which can be made more conspicuous by adding the ordinal numeral eerste'first'. Consider the ambiguous example in (549a). If the van-PP is dependent on the story noun, as in (549b), we are not dealing with the first book that Jan has ever read; it merely states that it is the first time Jan has read a book by Huizinga. If the van-PP is independent of the story noun, as in (549b'), the resulting sentence is slightly marked, but it is clear that the first book read by Jan was written by Huizinga.

    549
    a. Jan heeft zijn eerste boek van Huizinga gelezen.
      Jan has  his first book  of Huizinga  read
      'Jan (has) read his first book by Huizinga.'
    b. Jan heeft [NP zijn eerste boek [PP van Huizinga]Agent] gelezen.
    b'. ? Jan heeft [NP zijn eerste boek] [PP van Huizinga]Adv gelezen.

    Sentence (550a) does not allow this type of ambiguity: it only allows an interpretation associated with (550b) according to which Jan has burnt a book written by Huizinga for the first time; the reading associated with (550b'), according to which the very first book that Jan ever burnt was a book by Huizinga, is not available. Another (less likely) reading of (550a) is that Jan is planning to burn a number of books by Huizinga, and that he has now burnt the first one, but on this reading, too, the van-PP would function as a complement of the noun.

    550
    a. Jan heeft zijn eerste boek van Huizinga verbrand.
      Jan  has  his first book  of Huizinga  burnt
      'Jan (has) burnt his first book by/from Huizinga.'
    b. Jan heeft [NP zijn eerste boek [PP van Huizinga]Agent] verbrand.
    b'. * Jan heeft [NP zijn eerste boek] [PP van Huizinga]Adv verbrand.
    [+]  2.  Negation

    We have seen earlier that insertion of the negator niet'not' can also be used to disambiguate these sentences on the assumption that it can only be placed between the noun and an independent over-PP. We therefore expect niet to be acceptable in this position in (551a) on the reading associated with the structures in (551b'). This expectation seems to be borne out: (551a) can only be interpreted in such a way that the book in question is the very first book about the Middle Ages I ever read, that is, with the PP as independent constituent. Observe, however, that the use of sentences of this kind is somewhat restricted: they are only fully acceptable in contrastive contexts and leaving out the theme argument will degrade the result.

    551
    a. Ik heb mijn eerste boek over de middeleeuwenTheme niet van HuizingaAgent gelezen, maar van PleijAgent.
      I  have  my first book  about the Middle Ages  not of Huizinga  read  but  of Pleij
      'The first book about the Middle Ages I read wasnʼt by Huizinga but by Pleij.'
    b. * Ik heb [NP mijn eerste boek over de M.E. [PP niet van Huizinga]Agent] gelezen.
    b'. Ik heb [NP mijn eerste boek over de M.E.] niet [PP van Huizinga]Adv gelezen.

    The assumption that the negator niet'not' can only be placed between the noun and an independent over-PP also correctly predicts that (552a) is unacceptable, as only the structure with a dependent PP is available.

    552
    a. * Ik heb mijn eerste boek niet van Huizinga verbrand.
      I  have  my first book  not  of Huizinga  burnt
    b. * Ik heb [NP mijn eerste boek [PP niet van Huizinga]Agent] verbrand.
    b'. * Ik heb [NP mijn eerste boek] niet [PP van Huizinga]Adv verbrand.
    [+]  3.  Topicalization

    Topicalization can also be used to distinguish the two readings. Of course, since van-PPs can be dependent on the noun with both verb types, it does not come as a surprise that it is always possible to topicalize both the noun phrase and the PP. Since in Dutch only a single constituent can be placed before the verb in sentence initial position (the constituency test), we predict that only the constructions in the singly-primed examples are acceptable. This predication is borne out: sentence (553a) means that I read my first book by Huizinga, not that this was the first book I read.

    553
    a. Mijn eerste boek van Huizinga heb ik gelezen.
      my first book  of Huizinga  have  I  read
      'I (have) read my first book by Huizinga.'
    a'. [NP Mijn eerste boek [PP van Huizinga]Agent]i heb ik ti gelezen.
    a''. * [NP Mijn eerste boek]i [PP van Huizinga Adv]j heb ik titj gelezen.
    b. Mijn eerste boek van Huizinga heb ik verbrand.
      my first book  of Huizinga  have  I  burnt
      'I (have) burnt my first book by Huizinga.'
    b'. [NP Mijn eerste boek [PP van Huizinga]Agent]i heb ik ti verbrand.
    b''. * [NP Mijn eerste boek]i [PP van Huizinga Adv]j heb ik titj verbrand.

          Next, let us consider the possibilities for topicalizing the story noun only, leaving the PP in the original position. Here we would expect only the doubly-primed examples to yield acceptable results. The examples in (554) show that this prediction is borne out: example (554a) can only express that the book in question is the very first one I ever read. Example (554b) does not allow this type of topicalization, as the only structure that would allow this kind of topicalization (the doubly-primed one) is not available.

    554
    a. ? Mijn eerste boek heb ik van Huizinga gelezen.
      my first book  have  I  of Huizinga  read
      'The first book I read was by Huizinga.'
    a'. * [NP Mijn eerste boek]i heb ik [NPti [PP van Huizinga]Agent] gelezen.
    a''. [NP Mijn eerste boek]i heb ik ti [PP van Huizinga]Adv gelezen.
    b. * Mijn eerste boek heb ik van Huizinga verbrand.
      my first book  have  I  of Huizinga  burnt
    b'. * [NP Mijn eerste boek]i heb ik [NPti [PP van Huizinga]Agent ] verbrand.
    b''. * [NP Mijn eerste boek]i heb ik ti [PP van Huizinga]Adv verbrand.

          Finally, let us look at the possibilities for topicalizing the van-PP, and assume that only independent van-PPs can be topicalized. The relevant examples can be found in (555). Example (555a) indeed seems to have the interpretation associated with the independent use of the PP: it expresses that the very first book I read was about the Middle Ages. As expected, example (555b) is degraded due to the fact that the over-PP can only interpreted as a dependent of the story noun.

    555
    a. Van Huizinga heb ik mijn eerste boek gelezen.
      of Huizinga  have  I  my first book  read
      'I (have) read my first book by Huizinga.'
    a'. * [PP Van HuizingaAgent]i heb ik [NP mijn eerste boek ti] gelezen.
    a''. [PP Van Huizinga Adv]i heb ik [NP mijn eerste boek] ti gelezen.
    b. ?? Van Huizinga heb ik mijn eerste boek verbrand.
      of Huizinga  have  I  my first book  burnt
      'I (have) burnt my first book by Huizinga.'
    b'. ?? [Van HuizingaAgent]i heb ik [NP mijn eerste boek ti] verbrand.
    b''. * [Van Huizinga Adv]i heb ik [NP mijn eerste boek] ti verbrand.

    It should be noted, however, that it is possible to force the reading associated with the structure in (555a'), according to which the book in question is not the very first I read, but the first I read by Huizinga, by adding an adverb like onlangs'recently'. Clearly, this cannot be accounted for by assuming that only PPs functioning as independent constituents can be topicalized; we should therefore assume the same thing to be possible with complements of story nouns.

    556
    a. Van Huizinga heb ik onlangs mijn eerste boek gelezen.
      of Huizinga  have  I  recently  my first book  read
      'I (have) recently read my first book by Huizinga.'
    b. [PP Van HuizingaAgent]i heb ik onlangs [NP mijn eerste boek ti] gelezen.
    b'. [PP Van Huizinga Adv]i heb ik onlangs [NP mijn eerste boek] ti gelezen.
    [+]  C.  Story nouns with verbs like zien'to see' and vertalen'to translate'

    This subsection discusses constructions with verbs such as zien'to see' or vertalen'to translate' to see whether there are reasons for assuming that these, too, can be combined with an independent over- or van-PP. The answer will be negative; we will show that the structure in the doubly-primed examples are excluded.

    557
    a. Ik heb een film over Nixon gezien.
      I  have  a film  about Nixon  seen
      'Iʼve seen a film about Nixon.'
    a'. Ik heb [een film [over Nixon]Theme] gezien.
    a''. *? Ik heb [een film] [over Nixon]Adv gezien.
    b. Ik heb een boek van Huizinga vertaald.
      I  have  a book  of Huizinga  translated
      'Iʼve translated a book by Huizinga.'
    b'. Ik heb [een boek [van Huizinga]Agent] vertaald.
    b''. * Ik heb [een boek] [van Huizinga]Adv vertaald.

    The first reason for this conclusion is that adding the ordinal numeral eerste does not create the by now familiar ambiguity. The most plausible interpretation of example (558a) is that in which I have just for the first time seen a film about Nixon, or, alternative, that I watched the first film about Nixon that I ever purchased; in both cases we are dealing with a theme argument of the noun. Something similar holds for the construction in example (558b): the book in question is not the first one I ever translated, but the first one by Huizinga.

    558
    a. Ik heb mijn eerste film over Nixon gezien.
      I  have  my first film  about Nixon  seen
      'Iʼve seen my first film about Nixon.'
    a'. Ik heb [mijn eerste film [over Nixon]Theme] gezien.
    a''. *? Ik heb [mijn eerste film] [over Nixon]Adv gezien.
    b. Ik heb mijn eerste boek van Huizinga vertaald (in het Engels).
      I  have  my first book  of Huizinga  translated  into the English
      'Iʼve translated my first book by Huizinga (into English).'
    b'. Ik heb [mijn eerste boek [van Huizinga]Agent] vertaald.
    b''. * Ik heb [mijn eerste boek] [van Huizinga]Adv vertaald.

          On the basis of these facts, we expect placement of the negator niet in between the story noun and the over-PP to yield an unacceptable result. The examples in (559) show that this expectation is indeed borne out: both sentences are semantically ill-formed.

    559
    a. * Ik heb een film niet over NixonTheme gezien (maar over JFK).
      I  have  a film  not about Nixon  seen   but about JFK
    b. * Ik heb een boek niet van HuizingaAgent vertaald (maar van Pleij).
      I  have  a book  not of Huizinga  translated   but of Pleij

    Topicalization, on the other hand, provides more equivocal evidence. The fact illustrated by (560a) and (561a) that topicalization of the entire noun phrase is possible again confirms that the primed structures in (558) are available, but evidence in favor of the claim that the doubly-primed structures are ungrammatical is less firm: the claim is supported by the fact illustrated by the (b)-examples that topicalization of the noun phrase in isolation is impossible, but contradicted by the fact illustrated by the (c)-examples that topicalization of the van-PP in isolation is allowed. Note that the results of the negation and topicalization tests change if we use a determiner like zoʼn'such a' or een dergelijke'a similar'; see Subsection IB for discussion.

    560
    a. Mijn eerste film over Nixon heb ik pas gisteren gezien.
      my first film about Nixon have  I  only yesterday  seen
    b. * Mijn eerste film heb ik over NixonTheme gezien (niet over JFK).
      my first film  have  I  about Nixon  seen   not about JFK
    c. Over Nixon heb ik pas gisteren mijn eerste film gezien.
      about Nixon  have  I  only yesterday  my first film seen
    561
    a. Mijn eerste boek van HuizingaAgent heb ik in 1998 vertaald.
      my first book of Huizinga  have  I  in 1998 translated
    b. * Mijn eerste boek heb ik van HuizingaAgent vertaald (niet van Pleij).
      my first book  have  I  of Huizinga  translated   not of Pleij
    c. Van HuizingaAgent heb ik mijn eerste boek in 1998 vertaald.
      of Huizinga have  I  my first book in 1998 translated
    [+]  D.  Conclusion

    It seems that we may tentatively conclude on the basis of the preceding subsections that there is no reason for assuming that sentences with the verbs zien'to see' and vertalen'to translate' are structurally different from affective verbs like verbranden'to burn': they only allow the reading with the PP inside the noun phrase. Whereas there is clear syntactic and semantic evidence for assuming that communication verbs like schrijven'to write' may occur in two different syntactic frames, such evidence is lacking in the case of verbs like zien and vertalen. The fact that the latter verbs allow topicalization of the van/over-PP therefore suggests that topicalization is not restricted to independent PPs, but can also apply to complements of story nouns. The test of PP-extraction may therefore still be a valid means of distinguishing between complements and adjuncts within the noun phrase.

    References:
    • Haan, Ger de1979Onafhankelijke PP-komplementen van nominaSpektator8330-339
    • Klein, M. & Toorn, M.C. van den1979Van NP-beperking to XP-beperking: een antwoord op Kooij en Wiers 1978De Nieuwe Taalgids7297-109
    • Klein, M. & Toorn, M.C. van den1979Van NP-beperking to XP-beperking: een antwoord op Kooij en Wiers 1978De Nieuwe Taalgids7297-109
    • Kooij, Jan & Wiers, Evelyn1977Vooropplaatsing van PPs in het NederlandsSpektator6445-449
    • Kooij, Jan & Wiers, Evelyn1978Vooropplaatsing, verplaatsingsregels en interne structuur van nominale groepenKooij, Jan (ed.)Aspekten van woordvolgorde in het NederlandsLeidenVakgroep Nederlandse Taal- & Letterkunde105-143
    About this article
    Author(s): Hans Broekhuis, Evelien Keizer
    Category: Dutch Syntax
    Keywords: Constituent NegationNegationTopicalization
    Version: 1.0, December 2015
    Version history:
    version editor(s) date remarks
    1.0 Hans Broekhuis, Evelien Keizer December 2015
    report errorprintcite
    Jump to:
    X