• Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
5.2.2.2.Subject raising infinitivals
quickinfo

The infinitival clauses with te-infinitives discussed in this section differ from the ones discussed in Section 5.2.2.1 in that they do not involve the implied subject PRO, but take a lexical subject which is subsequently raised to the subject position of the matrix clause in order to receive nominative case. The difference between control and subject raising infinitivals is indicated schematically in (469).

469
a. [NPi Vfinite [infinitival clause PROi ... te Vinf ...]].
b. [NPi Vfinite [infinitival clause ti ... te Vinf ...]].

Typical examples of verbs triggering subject raising are the evidential modal verbs in (470a&b), but there are also verbs that occur incidentally in subject raising constructions, like dreigen and beloven in (470c).

470
Subject Raising verbs
a. Modal verbs: blijken'to turn out', lijken'to appear', schijnen'to seem'
b. Modal verbs (formal): dunken'to seem/be of the opinion', heten'to call/count oneself', toeschijnen'to seem', voorkomen'to appear'
c. Other: dreigen'to threaten' and beloven'to promise'

This section is organized as follows, subsection I starts by introducing the term subject raising and provides some general syntactic properties of subject raising constructions, subsection II continues with a more detailed discussion of the subject raising verbs in (470), subsection III concludes with the discussion of a more restricted type of expression, which we will refer to as passive subject raising constructions.

readmore
[+]  I.  General properties of subject raising constructions

Subsection A shows that subject raising constructions can be distinguished from control constructions by means of pronominalization, subsection B discusses two different analyses of subject raising verbs, namely, as main or non-main verbs; we will show that, in keeping with our definition of non-main verbs (verbs lacking argument structure), we have to do with main verbs, subsection C concludes by pointing out a number of characteristic syntactic properties of subject raising constructions.

[+]  A.  Subject Raising versus control infinitivals: pronominalization

Consider the examples in (471). Example (471a) shows that blijken'to turn out' is a monadic verb that may take a finite subject clause, which is introduced by the anticipatory pronounhet'it' (we ignore for the moment that in some cases blijken may also take an indirect object); that the clause functions as a subject is clear from the fact illustrated in (471b) that substitution of a lexical DP/referential pronoun for the pronoun het leads to ungrammaticality.

471
a. Het bleek [dat Jan een auto gekocht had].
  it  turned.out   that  Jan  a car  bought  had
  'It turned out that Jan had bought a car.'
b. * Marie/Zij bleek [dat Jan een auto gekocht had].
  Marie/she  turned.out   that  Jan  a car  bought  had

At first sight, the primeless examples in (472) seem to contradict the claim that blijken is monadic. The noun phrases Jan and Jan en Marie clearly function as the subjects of these sentences, as is clear from the fact that they agree in number with the verb blijken. There are nevertheless reasons for assuming that these nominative subjects are not arguments of the modal verb blijken but of the infinitival verb embedded under it. The most important reason for assuming this is that it is not possible to pronominalize the italicized parts of the examples in (472) while maintaining the nominative DP; pronominalization also requires the subject of the infinitival to be omitted. This is shown in the primed examples in (472).

472
a. Jan bleek een auto gekocht te hebben.
  Jan turned.out  a car  bought  to have
  'Jan turned out to have bought a car.'
a'. Dat bleek.
  that  turned.out
a''. * Jan bleek dat.
  Jan turned.out  that
b. Jan en Marie bleken een auto gekocht te hebben.
  Jan and Marie  turned.out  a car  bought  to have
  'Jan and Marie turned out to have bought a car.'
b'. Dat bleek.
  that  turned.out
b''. * Jan en Marie bleken dat.
  Jan and Marie  turned.out  that

In this respect, subject raising constructions conspicuously differ from control constructions such as (473a), in which pronominalization of the infinitival clause cannot affect the nominative subject of the matrix clause, as shown by (473b).

473
a. Jani probeert [PROi dat boek te lezen].
  Jan  tries  that book  to read
  'Jan is trying to read that book.'
b. Jan probeert dat. / *Dat probeert.
  Jan tries  that    that tries

The contrast between the examples in (472) and (473) suggests that the nominative noun phrases Jan and Jan en Marie in (472) originate as part of the infinitival clause and are raised to the subject positions of the matrix clauses, as in the representations in (474).

474
a. Jani bleek [ti een auto gekocht te hebben].
b. [Jan en Marie]i bleken [ti een auto gekocht te hebben].

The movement is normally taken to be an instantiation of NP-movement, which implies that the motivation of this movement is the need of the noun phrase to be assigned case: the noun phrase cannot be assigned case from within the infinitival clause, for which reason it is raised to the subject position of the sentence where it can be assigned nominative case.

[+]  B.  The status of the subject raising verb: main or non-main verb?

It seems that the standard analysis in (474) has no implications for the status of the subject raising verb: it seems compatible with the traditional claim that modal verbs like blijken'to turn out', schijnen'to seem' and lijken'to appear' are non-main verbs, but also with the claim that they are main verbs. In fact, it is not immediately clear whether the two positions are really different from a syntactic point of view, given that they both maintain that the subject of the sentence, Jan/Jan en Marie, is an argument of the predicate in the te-infinitival. However, the two claims do make different predictions concerning the examples in (475), at least if we adopt our earlier definition of non-main verbs as verbs that do not assign thematic roles. Example (475a) shows that lijken'to appear' is a dyadic verb that selects an experiencer argument in addition to a clausal subject. If the subject raising construction in (475b) involves a non-main verb, and if non-main verbs are not able to select arguments, we wrongly predict that the experiencer argument cannot be realized in this construction. This implies that, according to our definition of non-main verbs, modal verbs like blijken, schijnen and lijken are also main verbs in subject raising constructions.

475
a. Het lijkt mij [dat Jan goed past in onze groep].
  it  appears  me   that  Jan  well  fits  in our team
  'It appears to me that Jan will fit well in our team.'
b. Jani lijkt mij [ti goed in onze groep te passen].
  Jan  appears  me  well  in our team  to fit
  'Jan appears to me to fit well in our team.'

The subject raising analysis of infinitival constructions with blijken, schijnen and lijken is essentially identical to the analysis of examples such as (476), in which these verbs take a complementive; these constructions are traditionally analyzed as copular constructions. The primed examples show that the nominative noun phrase is generated as the logical subject of an embedded predicate, with which it forms a so-called small clause, and is subsequently raised to the subject position in order to receive nominative case.

476
a. Jan bleek/leek/scheen erg aardig.
  Jan turned.out/appeared/seemed  very nice
  'Jan turned out/appeared/seemed very nice.'
a'. Jani bleek/leek/scheen [SCti erg aardig].
b. Jan bleek/leek/scheen een goede vriend.
  Jan turned.out/appeared/seemed  a good friend
  'Jan turned out/appeared/seemed a good friend.'
b'. Jani bleek/leek/scheen [SCti een goede vriend].

The main difference between subject raising and complementive constructions is the status of the complement of the verb; is it an infinitival clause (that is a verbal predicative phrase) or a small clause (a predicate of some other category)? It therefore does not come as a surprise that examples such as (476) alternate with the those in (477), which contain an infinitival copular construction.

477
a. Jan bleek/leek/scheen erg aardig te zijn.
  Jan turned.out/appeared/seemed  very nice to be
  'Jan turned out/appeared/seemed to be very nice.'
a'. Jani bleek/leek/scheen [Clauseti erg aardig te zijn].
b. Jan bleek/leek/scheen een goede vriend te zijn.
  Jan turned.out/appeared/seemed  a good friend  to be
  'Jan turned out/appeared/seemed to be a good friend.'
b'. Jani bleek/leek/scheen [Clauseti een goede vriend te zijn].

On this view there is no need for assuming that blijken, schijnen and lijken are ambiguous: we are not dealing with a set of modal and a set of copular verbs, but simply with a single category that takes a predicative complement that may either have the form of an infinitival clause or of a small clause; in both cases the subject of the predicate is raised to the subject position of the clause headed by the modal verb in order to receive nominative case.

[+]  C.  Syntactic properties of subject raising constructions

The conclusion from subsection B that subject raising verbs are main verbs raises several questions, which will be discussed in the following subsections.

[+]  1.  Om + te-infinitivals are excluded

Subject raising verbs differ from control verbs in that they do not take om + te-infinitivals. The unacceptability of the subject raising construction in (478b) is easy to account for, given that Section 5.2.2.1 has independently established that om + te-infinitivals are syntactic islands for movement, and can therefore be assumed to block subject raising. It is, however, less clear why (478c) is unacceptable, especially since (471a) has shown that similar constructions are possible with finite clauses; this unacceptability is possibly due to the fact that there is no suitable controller available for the implied subject PRO (cf. Bennis & Hoekstra 1989a).

478
a. Jani schijnt [ti de boeken gestolen te hebben].
  Jan seems  the books  stolen  to have
  'Jan seems to have stolen the books.'
b. * Jani schijnt [om ti de boeken gestolen te hebben].
  Jan seems  comp  the books  stolen  to have
c. * Het schijnt [om PRO de boeken gestolen te hebben].
  it  seems  comp  the books  stolen  to have

Such an account of the unacceptability of (478c) would leave unexplained, however, why the (c)-example in the parallel set of examples in (479) is unacceptable as well, given that the experiencer object me of lijken'to appear' could in principle function as a controller for PRO. We will not pursue this issue here and leave it for future research.

479
a. Jani lijkt me [ti de boeken gestolen te hebben].
  Jan appears  me   the books  stolen  to have
  'Jan appears to me to have stolen the books.'
b. * Jani lijkt me [om ti de boeken gestolen te hebben].
  Jan appears  me  comp  the books  stolen  to have
c. * Het lijkt me [om PRO de boeken gestolen te hebben].
  it  appears  me  comp  the books  stolen  to have
[+]  2.  The complement is a transparent infinitival (verb clustering and IPP)

The examples in (478a&b) in the previous subsection show that infinitival clauses of subject raising constructions must be transparent for NP-movement. This is consistent with the fact that such clauses are transparent infinitivals in the sense defined in Section 4.4.3: subject raising constructions exhibit verb clustering (and thus require the embedded infinitival clause to be split), and the te-infinitive seems to trigger the infinitivus-pro-participio (IPP) effect on the matrix verb in perfect-tense constructions. The former can be illustrated by the contrast between the two examples in (480).

480
a. dat Jan de boeken naar Groningen schijnt te sturen.
  that  Jan  the books  to Groningen  seems  to send
  'that Jan seems to send the books to Groningen.'
b. * dat Jan schijnt de boeken naar Groningen te sturen.
  that  Jan  seems  the books  to Groningen  to send

That subject raising constructions exhibit the IPP-effect is less easy to illustrate given that many speakers tend to object to perfect-tense constructions with evidential modal verbs; see Haeseryn et al. (1997:958) and also Schmid (2005:27), who claims that subject raising constructions tend to resist perfectivization cross-linguistically. Nevertheless, it seems that some speakers do at least marginally accept perfect-tense constructions such as (481), and then always prefer the IPP-effect; replacement of the infinitives schijnen, lijken and blijken in (481) by the corresponding participial forms geschenen, geleken and gebleken indeed greatly worsens the results; see Reuland (1983: Section 3.2) and Rutten (1991:70).

481
a. % dat Jan de boeken naar Groningen heeft schijnen te sturen.
  that  Jan the books  to Groningen  has  seem  to send
  'that Jan has seemed to send the books to Groningen.'
b. % dat Jan de boeken naar Groningen heeft lijken te sturen.
  that  Jan the books  to Groningen  has  appear  to send
  'that Jan has appeared to send the books to Groningen.'
c. % dat Jan de boeken naar Groningen heeft blijken te sturen.
  that  Jan the books  to Groningen  has  turn.out  to send
  'that Jan has turned out to send the books to Groningen.'

Note in this connection that Van der Horst (2008:1464&1796) claims that constructions with schijnen have exhibited the IPP-effect already since the 18th century, which he illustrates with a single example only. However, his claim can easily be substantiated by means of a Google Book search on the string [heeft schijnen te]. Van der Horst (2008:1769) also provides a number of recent IPP-cases with blijken, and a Google Book search on the string [heeft blijken te] again provides a number of additional cases. There are, however, also a number of relevant hits for [heeft geschenen te] and [heeft gebleken te]. Van der Horst does not discuss cases with the verb lijken, and a Google Book search on the strings [heeft lijken/geleken te] did not result in any relevant hits either, but see Haegeman (2006) for the claim that lijken does occur in the perfect tense. The results of our searches are given in Table (482); the reported results were checked manually and exclude hits from linguistic sources.

482
Google Book search (1/13/2013) on the string [heeft modalinf/part te]
infinitive participle
schijnen'to seem' 12 2
blijken'to turn out' 11 6
lijken'to appear' 0 0

The results in (482) are, of course, based on older written sources and are certainly not representative of present-day use. Unfortunately, the results of our Google searches on the strings [heeft modalinf/part te] are far too polluted by irrelevant cases (often machine translations from English) to allow anything enlightening to be said about the frequency on the internet of genuine cases of constructions such as (481) with and without IPP, apart from the fact that the numbers are low anyway. We therefore have to leave this issue to future research, and provisionally assume that, insofar as perfect-tense forms of subject raising constructions are possible at all, they preferably exhibit the IPP-effect.

[+]  3.  Subject Raising verbs are unaccusative

A more technical question raised by assuming that subject raising verbs are main verbs concerns the argumental status of the infinitival clause: Is it an internal or an external argument of the modal verb, that is, are we dealing with unaccusative verbs?
      The unaccusative analysis seems a plausible one; because the subject of the infinitival clause uncontroversially surfaces as the nominative subject of the matrix clause, it seems unlikely that the infinitival clause is generated as the external argument of the matrix verb given that such arguments normally must surface as the subject of active constructions—this would make subject raising impossible. If the infinitival clause is generated as an internal argument of the verb, there is no external argument and we may conclude that, as a result of this, the subject of the infinitival clause is able to raise to the subject position of the higher clause.
      That we are dealing with unaccusative verbs is also supported by the fact that blijken takes zijn in the perfect tense (in non-IPP-contexts): Dat is/*heeft gebleken'That has turned out'; selection of the perfect auxiliary zijn is a sufficient condition for assuming unaccusative status. The complementive constructions in (483) show that schijnen and lijken do not allow zijn in the perfect tense; that these verbs seem to prefer hebben is, however, not a problem given that the selection of zijn is not a necessary condition for assuming unaccusative status; cf. Section 2.1.2.

483
a. Jan heeft/*is me altijd aardig geleken.
  Jan has/is  me always  kind  seemed
  'Jan has always seemed kind to me.'
b. Jan ?heeft/*is altijd aardig geschenen.
  Jan   has/is  always  nice  appeared
  'Jan has always appeared kind.'
[+]  4.  Passivization

The conclusion that subject raising verbs are unaccusative correctly predicts that such verbs do not allow impersonal passivization. This is illustrated in (484) for the verb lijken in the three syntactic contexts in which it may occur. The reason why the nominative subjects cannot be suppressed in the primed examples is that they are not arguments of the passivized verb but originate as arguments of the complements of this verb; for convenience, the (split) complements are in italics in the primeless examples.

484
a. Het lijkt me dat Jan morgen komt.
finite subject clause
  it  appears  me  that  Jan  tomorrow  comes
  'It appears to me that Jan will come tomorrow.'
a'. * Er wordt me geleken dat Jan morgen komt.
  there  is  me appeared  that  Jan tomorrow  comes
b. Jan lijkt me morgen te komen.
subject raising
  Jan  appears  me  tomorrow  to come
  'Jan appears to me to come tomorrow.'
b'. * Er wordt me geleken morgen te komen.
  there  is  me  appeared  tomorrow  to come
c. Jan lijkt me geschikt voor die baan.
complementive
  Jan  appears  me suitable  for that job
  'Jan appears suitable for that job to me.'
c'. * Er wordt me geschikt geleken voor die baan.
  there  is  me suitable  appeared  for that job

The (b)-examples in (485) show that passivization of the embedded infinitival clause is possible; the (a)-examples are simply given for comparison. As predicted by the subject raising analysis, passivization of the infinitival clause also affects the nominative subject of the subject raising construction as a whole; the internal argument of the infinitival verb, de auto, surfaces as the nominative subject of the construction as a whole, while the subject of the active construction, Jan, is suppressed; in short, it is the derived subject in (485a') that becomes the nominative subject of the entire construction.

485
a. Het lijkt me dat Jan de auto repareert.
finite subject clause
  it  appears  me that  Jan  the car  repairs
  'It appears to me that Jan is repairing the car.'
a'. Het lijkt me dat de auto gerepareerd wordt.
  it  appears  me that  the car  repaired  is
  'It appears to me that the car is being repaired.'
b. Jan lijkt me de auto te repareren.
subject raising
  Jan  appears  me the car  to repair
  'Jan appears to me to repair the car.'
b'. De auto lijkt me gerepareerd te worden.
  the car  appears  me repaired  to be
  'The car appears to me to be repaired.'

Finally, consider the examples in (486) adapted from Bennis & Hoekstra (1989c:172); the judgments hold only for speakers that allow passivization of the idiomatic expression de strijdbijl begraven'to bury the hatchet/to make peace'. The fact that the idiomatic reading is preserved in (486b') can be taken as in favor of the claim that the noun phrase de strijdbijl is base-generated as part of the infinitival clause: since phrasal idioms are listed in the lexicon, the expression de strijdbijl begraven must be inserted into the structure as a unit.

486
a. Het schijnt dat Jan en Marie de strijdbijl hebben begraven.
  it  seems  that  Jan and Marie  the hatchet  have  buried
  'It seems that Jan and Marie have buried the hatchet.'
a'. Jan en Marie schijnen de strijdbijl te hebben begraven.
  Jan and Marie  seem  the hatchet  to have  buried
  'Jan and Marie seem to have buried the hatchet.'
b. Het schijnt dat de strijdbijl begraven is.
  it  seems  that  the hatchet  buried  has.been
  'It seems that has been buried the hatchet.'
b'. De strijdbijl schijnt begraven te zijn.
  the hatchet  seems buried  to have.been
  'The hatchet seems to have been buried.'
[+]  5.  Subject raising is excluded in nominalizations

Subject raising requires the te-infinitival to be a complement of a verb; the primed examples in (487) show that whereas non-raising constructions such as (487a) have nominal counterparts, subject raising constructions such as (487b) have not.

487
a. het schijnt [dat Jan ziek is].
  it  seems   that  Jan ill  is
  'It seems that Jan is ill.'
b. Jani schijnt [ti ziek te zijn].
  Jan seems  ill  to
  'Jan seems to be ill.'
a'. de schijn [dat Jan ziek is]
  the appearance   that Jan ill is
  'the pretense that Jan is ill'
b'. * Jansi schijn [ti ziek te zijn]
  Janʼs  appearance  ill  to be

This suggests that te-infinitival complements of nouns differ from those of verbs in that they are not transparent. This is in line with Koster's (1984b) claim, discussed in Section 5.2.2.1 that te-infinitival complements of nouns do not involve obligatory control either.

[+]  6.  Conclusion

The facts discussed in the previous subsections conclusively show that subjects of subject raising constructions cannot be analyzed as arguments of the subject raising verb but originate as arguments of the embedded infinitival verb, subject raising occurs out of te-infinitival complements of certain unaccusative verbs (but not of their corresponding nominalizations)

[+]  II.  Subject raising verbs

Subject raising verb normally have a modal meaning. This is especially clear for the modal verbs blijken'to turn out', lijken'to appear', and schijnen'to seem' in (470a), which are traditionally analyzed as (semi-)auxiliaries in this context, but it also holds for verbs like beloven'to promise' and dreigen'to threaten' in (470c), which are used more incidentally in this construction. The following subsections briefly discuss these verbs in more detail, subsection A begins by having a closer look at the modal verbs blijken, lijken and schijnen, subsection B discusses the verbs in (470c) while Subsection C concludes with the more formal modal verbs in (470b) as well as a number of other potential cases from the formal register.

[+]  A.  The verbs blijken, schijnen and lijken

Adopting the categorization of modality proposed by Palmer (2001), which is discussed more extensively in Section 5.2.3.2, sub III, we may classify verbs like blijken'to turn out', lijken'to appear', and schijnen'to seem' in (488) semantically as evidential modals, in the sense that they can be used to indicate what kind of evidence there is in favor of the truth of a certain proposition p: see Van Bruggen (1980/1), Haeseryn et al. (1997:1007-8), Vliegen (2011) and Koring (2013) for discussion. The verb blijken suggests that there is conclusive evidence to conclude that p is true, in the sense that on the basis of this evidence most people would conclude that p is true. The verb lijken expresses that there is evidence in support of p but that the evidence is not yet conclusive; on the basis of the evidence one can only provisionally assume that p is true. The verb schijnen, finally, expresses that there is no identifiable evidence that supports p; the evidence may or may not exist—we are dealing with hearsay/rumors.

488
a. Uit zijn verklaring blijkt [dat Jan de dader is].
conclusive
  from his statement  turns.out   that  Jan the perpetrator  is
  'His statement clearly shows that Jan is the perpetrator.'
b. Het lijkt mij/haar [dat Jan de dader is].
not yet conclusive
  it  appears  me/her   that  Jan the perpetrator  is
  'It appears to me/her that Jan is the perpetrator.'
c. Het schijnt [dat Jan de dader is].
hearsay
  it  seems   that  Jan the perpetrator  is
  'It seems that Jan is the perpetrator.'
[+]  1.  The verb blijken'to turn out'

The verb blijken expresses that there is factual evidence in support of the proposition expressed by the argument clause. Use of this verb further suggests that the truth of the proposition can at least be intersubjectively established on the basis of the evidence available, that is, most people who consider this evidence carefully would come to the same conclusion. Example (489a) shows that the nature of the factual evidence submitted can be specified by means of an adverbial uit-PP if the clause is finite, but not in the corresponding subject raising and complementive constructions in (489b&c); the latter examples nevertheless imply that the truth of the proposition expressed by the infinitival/small clause can be intersubjectively established. Adverbial uit-PPs of this type are normally not found with the verbs lijken and schijnen; see Table 3 in Vliegen (2011).

489
a. Uit zijn verklaring blijkt [dat Jan de dader is].
  from his statement  turns.out   that  Jan the perpetrator  is
  'His statement clearly shows that Jan is the perpetrator.'
b. Jan blijkt (*uit zijn verklaring) [ti de dader te zijn].
  Jan turns.out     from his statement  the perpetrator  to be
c. Jan blijkt (*uit zijn verklaring) [SCti de dader].
  Jan turns.out     from his statement  the perpetrator

Note in passing that examples such as (489) are perfectly fine if the preposition uit is replaced by volgens'according to': this may be due to the fact that the complement of the volgens-PP does not refer to the evidence on which the speakers bases his judgment of the truth of the proposition, but to the "judgment" provided by some source. While example (489) expresses that the speaker concludes from Jans statement that Jan is the perpetrator, an example like Volgens zijn verklaring blijkt dat Jan de dader is attributes this conclusion to Jan himself.
      It seems often implied that there is a specific set of individuals who have drawn the conclusion from the available evidence. With a finite complement clause the person(s) responsible for the conclusion can be expressed by means of a dative object (often the first person, plural pronoun ons'us'), which the literature normally refers to as the experiencer. The verb blijken should therefore be considered a nom-dat (dyadic unaccusative) verb. The addition of an experiencer leads to a degraded result in the corresponding subject raising and complementive constructions.

490
a. Er is ons gebleken [dat Jan de dader is].
  there  is  us  turned.out   that  Jan the perpetrator  is
  'We have concluded that Jan is the perpetrator.'
b. Jan bleek (*ons) [ti de dader te zijn].
  Jan  turned.out       us  the perpetrator to be
  'Jan turned out to be the perpetrator.'
c. Jan bleek (*ons) [SCti de dader].
  Jan turned.out     us  the perpetrator
  'Jan turned out to be the perpetrator.'

It should be noted that the use of an experiencer object is limited even in the case of finite argument clauses: it seems easily possible in perfect-tense constructions but is generally rejected in simple past/present constructions. The contrast is also clear from our Google search (31/1/2014): whereas the string [er is ons gebleken] resulted in 52 hits, the strings [er blijkt/bleek ons dat] resulted in no more than 9 relevant hits (all from very formal texts).

[+]  2.  The verb lijken'to appear'

The verb lijken indicates that the claim that the proposition expressed by the argument clause in (488a) is based on unmentioned evidence available; we are in a sense dealing with a subjective assessment of the evidence by a specific set of individuals, which includes the speaker by default. Example (491a) shows, however, that this set can also be made explicit by means of an optional experiencer object, in which case the default reading can readily be cancelled. The availability of an experiencer object shows that, like blijken, the verb lijken should be considered a nom-dat (dyadic unaccusative) verb. However, lijken differs from blijken in that the experiencer may also appear in the corresponding subject raising and complementive constructions in (491b&c).

491
a. Het lijkt mij/haar [dat Jan de dader is].
  it  appears  me/her   that  Jan the perpetrator  is
  'It appears to me/her that Jan is the perpetrator.'
b. Jani lijkt mij/haar [ti de dader te zijn].
  it  appears  me/her  the perpetrator  to be
  'Jan appears to me/her to be the perpetrator.'
c. Jani lijkt mij/haar [SCti de dader].
  it  appears  me/her  the perpetrator
  'Jan appears to be the perpetrator to me/her.'

      It seems that lijken differs from the other two verbs in that it can easily take a finite clause introduced by the linking element (als)of'as if'; the judgments on examples (492b) with schijnen vary from speaker to speaker, which is indicated by the percentage sign; we will briefly return to this issue in Subsection 4.

492
a. Het lijkt alsof Jan de dader is.
  it  appears  as.if  Jan the perpetrator  is
  'It appears as if Jan is the perpetrator.'
b. Het %schijnt/*blijkt alsof Jan de dader is.
  it    seems/turns.out  as.if  Jan  the perpetrator  is

This claim that modal lijken can be supplemented by an alsof-complement may be apparent, however, given that the verb lijken also occurs as a PO-verb with the meaning "to resemble/look like"; cf. example (493a). Since Section 2.3.1, sub VI, has shown that anticipatory pronominal PPs can often be omitted, it seems plausible to assume that example (492a) is a shorter form of example (493b) and thus does not involve the modal verb lijken.

493
a. Jan lijkt op zijn vader.
  Jan resembles  on his father
  'Jan resembles his father.'
b. Het lijkt erop alsof Jan de dader is.
  it  looks  like.it  as.if  Jan the perpetrator  is
  'It looks like Jan is the perpetrator.'

Example (494a), on the other hand, shows that (492a) can be extended with an experiencer object; the fact illustrated in (494b) that the experiencer and the anticipatory pronominal PP cannot co-occur therefore militates against the elision analysis. The bracketed numbers indicate the number of hits of our Google search (5/2/2013) for the search strings [het lijkt mij/me alsof] and [het lijkt mij/me erop alsof]. For completeness' sake notice that some speakers report that they consider example (494b) marked as well.

494
a. Het lijkt mij alsof Jan de dader is.
683
  it  appears  me  as.if  Jan the perpetrator  is
  'It appears to me like Jan is the perpetrator.'
b. * Het lijkt mij erop alsof Jan de dader is.
12
  it  appears  me  like.it  as.if  Jan the perpetrator  is

This leads to the conclusion that the evidential modal verb lijken can be supplemented by an alsof-complement after all.

[+]  3.  The verb schijnen'to seem'

An experiencer object is unlikely with the verb schijnen in examples such as (488c), and the same holds for the corresponding subject raising and complementive constructions in (495). The reason is that schijnen indicates that the truth of the proposition is based on rumors/hearsay; contrary to blijken and lijken, postulation of the truth of the proposition is not based on evidence available to any identifiable individual in the domain of discourse—it may in fact be entirely lacking.

495
a. Jani schijnt (*mij/*haar) [ti de dader te zijn].
  Jan  seems    me/her  the perpetrator  to be
  'Jan seems to be the perpetrator.'
b. Jani schijnt (*mij/*haar) [SCti de dader].
  Jan seems     me/her  the perpetrator
  'Jan seems to be the perpetrator.'

Moreover, the examples in (496) show that schijnen differs from blijken and lijken in that it does not readily allow pronominalization of its complement clause. It shows that evidence for claiming that the raising verb schijnen is a main verb is lacking; main verb status can only be argued on the basis of the assumption that schijnen belongs to the same class as blijken and lijken.

496
a. Dat blijkt later wel.
  that  turns.out  later  aff
  'That will become clear later.'
b. Dat lijkt me wel.
  that  appears  me  aff
  'That appears quite clear to me.'
c. * Dat schijnt later wel.
  that  seems  later  aff
[+]  4.  Conclusion

The three modal verbs blijken'to turn out', schijnen'to seem' and lijken'to appear' differ in the type of evidence available for testing the truth of the proposition expressed by the complement of the verb: blijken suggests that there is strong evidence on the basis of which the truth of the proposition can be objectively or intersubjectively established, the verb lijken suggests that the evidence is weaker in the sense that it is not conclusive and can be interpreted in a subjective manner, while the verb schijnen suggests that the nature of the evidence is unclear or may even be lacking; see Sanders & Spooren (1996) for experimental underpinning of these findings.
      It seems that speakers often have difficulties in giving judgments on examples with the modal verbs blijken, lijken and schijnen. One reason may be the interference of other constructions. We have seen, for example, that the dyadic modal verb lijken has a closely related counterpart that functions as a PO-verb with the meaning "to resemble/look like"; these verbs are diachronically derived from the same source and are still quite close in meaning. Furthermore, the situation in Dutch is rather special in that Dutch has two verbs, namely lijken and schijnen, where German simply has one verb, scheinen. This suggests that the meanings of these verbs are rather close which may give rise to a certain amount of confusion among speakers, especially since the introduction of lijken is quite recent—Vliegen (2011) suggests the 17th century.
      To conclude this discussion on blijken, lijken and schijnen, we want to note that they occur frequently in examples such as (497) where they are part of a phrase headed the preposition naar; the pronoun het'it' is often optional (especially with the verb blijken). Vliegen (2010) calls such naar-phrases parenthetic. Such an analysis may indeed be appropriate for cases such as Jan is (naar het schijnt) de beste leerling van zijn klas'As it seems, Jan is the best pupil of his class', but clearly wrong for examples such as (497) where this phrase is used in the first position of the sentence and must therefore be considered a constituent of the clause.

497
a. Naar het blijkt gaan ze naar de dierentuin.
  to  it  turns.out  go  they  to the zoo
  'It turns out that theyʼre going to the zoo.'
b. Naar het lijkt gaat het lukken.
  to  it  appears  goes  it  succeed
  'It appears that itʼll succeed.'
c. Naar het schijnt was ze elke dag dronken.
  to  it  seems  was she  every day  drunk
  'It seems that she was drunk every day.'

To our knowledge, examples such as (497), which can be quickly found on the internet by using the search string [naar (het) Vmodal], have not yet received a detailed analysis, and we therefore leave them to future research.

[+]  B.  The verbs dreigen'to threaten' and beloven'to promise'

In addition to the run-of-the-mill evidential modal verbs discussed in the previous subsection, there are various other verbs that may be found in subject raising constructions. This subsection discusses beloven'to promise' and dreigen'to threaten', and shows that these verbs have a number of special properties in their use as subject raising verbs; see also Verhagen (2005: Section 1.3.2) and Vliegen (2006) and references cited there. In order to set the stage, we will begin the discussion with the more regular uses of these verbs.

[+]  1.  Regular uses of beloven and dreigen

Beloven'to promise' and dreigen'to threaten' are generally used as verbs with an illocutionary meaning. The examples in (498) show that the illocutionary verb beloven is a triadic verb, which may select a noun phrase or a clause as its direct object. For our discussion in the following subsections, it is important to note that the complement in (498c) is an opaque infinitival in the sense of Section 4.4.3; it is in extraposed position and the infinitival verb does not trigger the IPP-effect, that is, the participle beloofd cannot be replaced by the corresponding infinitive beloven.

498
a. dat Jan (Marie) een cadeautje heeft beloofd.
  that  Jan   Marie  a present  has  promised
  'that Jan has promised Marie a present.'
b. dat Jan (Marie) heeft beloofd [dat hij morgen zou komen].
  that  Jan   Marie  has  promised   that  he  tomorrow  would  come
  'that Jan has promised Marie that he would come tomorrow.'
c. dat Jani (Marie) heeft beloofd [(om) PROi morgen te komen].
  that  Jan   Marie  has promised  comp  tomorrow  to come
  'that Jan has promised Marie to come tomorrow.'

The examples in (499) show that normally the illocutionary verb dreigen'to threaten' is an intransitive PO-verb, and that the complement of the PP can be either nominal or clausal; in the latter case the clause is optionally introduced by an anticipatory pronominal PP. For our discussion in the following subsections, it is again important to note that the complement in (499c) is not a transparent clause: it is in extraposed position and the infinitival verb does not trigger the IPP-effect, that is, the participle gedreigd cannot be replaced by the corresponding infinitive dreigen.

499
a. dat de directeur met collectief ontslag heeft gedreigd.
  that  the manager  with collective dismissal  has  threatened
  'that the manager has threatened collective dismissal.'
b. dat de directeur (ermee) heeft gedreigd [dat hij iedereen zal ontslaan].
  that  the manager  with.it  has  threatened   that  he  everyone  will dismiss
  'that the manager has threatened that he will dismiss everyone.'
c. dat de directeuri (ermee) heeft gedreigd [(om) PROi iedereen te ontslaan].
  that  the manager  with.it  has threatened  comp  everyone  to dismiss
  'that the manager has threatened to dismiss everyone.'
[+]  2.  The use of beloven and dreigen as subject raising verbs

Now that we have discussed the regular uses of beloven'to promise' and dreigen'to threaten', we can continue with their use as subject raising verbs in the examples in (500). That we are dealing with subject raising in these examples is clear from the fact that the inanimate noun phrases do not have the proper semantic properties to be assigned the agent roles of the illocutionary verbs beloven and dreigen.

500
a. Het boeki belooft [ti een succes te worden].
  the book  promises  a success  to become
  'The book promises to become a success.'
b. De boeki dreigt [ti op de vloer te vallen].
  the book  threatens  on the floor  to fall
  'The book threatens to fall on the floor.'

That the nominative subjects are not arguments of the verb beloven and dreigen receives more support from the fact illustrated in the primeless examples in (501) that the infinitival clause cannot be pronominalized without the subject of the sentence. In fact, the primed examples show that anticipatory elements het/ermee cannot be used to introduce the infinitival clause either.

501
a. * Het boek belooft het.
cf. Jan belooft het
  the book  promises  it
a'. * Het boeki belooft (het) [ti een succes te worden].
  the book  promises   it  a success  to become
b. * Het boek dreigt ermee.
cf. Jan dreigt ermee
  the book  threatens  with.it
b'. * Het boeki dreigt (ermee) [ti op de vloer te vallen].
  the book  threatens   with.it  on the floor  to fall

A third argument in favor of subject raising is that the complementizer om is prohibited: the ungrammaticality of the examples in (502) is as expected given that om + te-infinitivals are islands for movement and thus block the NP-movements indicated. Note that in this respect the modal verbs beloven and dreigen behave conspicuously different from the corresponding illocutionary verbs in (498c) and (499c), which readily allow om + te-infinitivals as their complements.

502
a. * Het boeki belooft [om ti een succes te worden].
  the book  promises  comp  a success  to become
  Intended reading: 'The book promises to become a success.'
b. * Het boeki dreigt [om ti op de vloer te vallen].
  the book  threatens  comp  on the floor  to fall
  Intended reading: 'The book threatens to fall on the floor.'

A fourth argument is that beloven and dreigen are like the run-of-the-mill subject raising verbs blijken'to turn out', schijnen'to seem' and lijken'to appear' discussed in Subsection A in that they are often not accepted in the perfect tense, but trigger the IPP-effect if speakers do accept it; this is clear from the fact that Barbiers (2006) marks (503) as unacceptable, whereas Van Dreumel and Coppen (2003) assign it a question mark to indicate that not all speakers consider it grammatical.

503
% Het heeft dreigen te stormen.
  it  has  threaten  to storm
'A gale has been threatening to blow up.'
[+]  3.  The meaning of beloven and dreigen in subject raising constructions

In subject raising constructions, beloven and dreigen assume an evidential or, perhaps, epistemic modal meaning; they express that the available evidence is sufficient for the speaker to conclude that the eventuality expressed by the infinitival clause will come to pass. The original illocutionary meaning of these verbs is lost: they no longer denote the illocutionary acts of promising or threatening but express, respectively, a positive and a negative evaluation held by the speaker of the eventuality expressed by the infinitival clause; cf. Verhagen (2005) and Vliegen (2006).
      That we are dealing with modal verbs is supported by the fact that the choice of present or past tense may affect the implications concerning the question as to whether the embedded proposition is actually realized; see Section 1.5.2 for similar observations regarding epistemic modals like moeten'must' and kunnen'may'. Consider the examples in (504). Example (504a) leaves entirely open whether Marie's promising career will actually lead to her being a great writer. Example (504b), on the other hand, strongly suggests that something unforeseen has occurred: Marie would have been a great writer if, e.g., she had not been killed in an accident.

504
a. Marie belooft een groot schrijver te worden.
  Marie promises  a great writer  to become
  'Marie promises to become a great author.'
b. Marie beloofde een groot schrijver te worden.
  Marie promised  a great writer  to become
  'Marie promised to become a great author.'

The reason for the negative implication in (504b) is pragmatic in nature and follows from Grice's (1975) maxim of quantity: if at the moment of speaking the speaker knows that Marie is already a great author, he can be more precise by simply using a present tense: Marie is een groot schrijver (geworden)'Marie is/has become a great author'. That we are dealing with pragmatics is also clear from the fact that any negative inference can be overridden by contextual information: for example, adding the adverbial phrase al vroeg'already early in her career' to example (504b) results in the positive implication that Marie is a great author at the moment of speaking; cf. Marie beloofde al vroeg een groot schrijver te worden'Already early in her career Marie promised to become a great author'. Of course, we do not only find this pragmatic effect in the case of beloven, but also (and perhaps more pervasively) in the case of dreigen.

[+]  4.  The predicate of the infinitival clause in subject raising constructions

The corpus investigation in Vliegen (2006) has shown that in the vast majority of cases the infinitival clauses embedded under modal beloven are copular constructions; cf. the examples in (505).

505
a. Jan belooft een goed mens te worden.
ambiguous
  Jan promises  a good person  to become
  'Jan promises (≈ makes a promise) to become a good person.'
  'Jan promises (≈ can be expected) to become a good person.'
b. Jan belooft het huis te kopen.
modal reading virtually excluded
  Jan promises  the house  to buy
  'Jan promises (≈ makes a promise) to buy the house.'

The predicate infinitival clauses embedded under modal dreigen, on the other hand, can be more varied; the examples in (506) are both fully acceptable in a modal reading.

506
a. Jan dreigt een slecht mens te worden.
modal reading preferred
  Jan threatens  a bad person  to become
  'Jan can be expected to become a bad person.'
b. Jan dreigt het huis te kopen.
ambiguous
  Jan threatens  the house  to buy
  'Jan threatens (≈ makes a threat) to buy the house.'
  'Jan can be expected to buy the house.'

The higher degree of productivity of the subject raising construction with modal dreigen may be related to the fact, also noted by Vliegen, that it arose earlier in the language than the corresponding construction with beloven, with the result that the illocutionary reading of the latter may be more prominent than that of the former.

[+]  5.  Ambiguous cases

The previous subsection has shown that constructions with beloven and dreigen can be ambiguous if the nominative subject is animate; cf. Bennis & Hoekstra (1989c:174-5). The verb beloven in examples such as (507a) can be interpreted as a control verb or as a subject raising verb, because there are no syntactic clues favoring one interpretation over the other. Of course, we abstract away from the fact that the (extra-)linguistic context may disambiguate (507a) by favoring a specific interpretation.

507
a. Jan belooft een goed mens te worden.
ambiguous
  Jan promises  a good person  to become
  'Jan promises to become a good person.'
b. Jani belooft [PROi een goed mens te worden].
control
b'. Jani belooft [ti een goed mens te worden].
subject raising

Example (507a) can be disambiguated by adding the complementizer om or (perhaps) by adding the anticipatory pronounhet'it' (which gives rise to a somewhat marked result here), as these additions both exclude the subject raising reading; for convenience, the elements originating inside the infinitival clause are italicized in (508).

508
a. Jan belooft om een goed mens te worden.
control only
  Jan promises  comp  a good person  to become
b. (?) Jan belooft het een goed mens te worden.
control only
  Jan promises  it  a good person  to become

Furthermore, the examples can also be disambiguated if they are used as embedded clauses. If the infinitival clause is in extraposed position, as in (509a), we normally interpret the construction as a control structure with an illocutionary verb (but see Subsection C for more discussion). If we are dealing with a split infinitival/verb clustering, as in (509b), we normally have to do with a subject raising construction with a modal verb (although Section 5.2.2.3 will show that we should be careful not to jump to conclusions in cases in which clause splitting seems to be possible).

509
a. dat Jan belooft een goed mens te worden.
control
  that  Jan promises  a good person  to become
b. dat Jan een goed mens belooft te worden.
subject raising
  that  Jan a good person  promises  to become

An additional way of disambiguating (507a) is the addition of an indirect object; example (510) does not allow a modal interpretation of the verb beloven. This is also clear from the fact illustrated in the (b)-examples that the infinitival clause cannot be split in embedded contexts The disambiguating effect of adding an indirect object indicates that control and subject raising verbs do not only differ in meaning but also in adicity. Note that we added a number sign to (510b') to indicate that, surprisingly, many speakers consider this example acceptable under a control reading; see Section 5.2.2.3 for discussion.

510
a. Jan belooft Marie een goed mens te worden.
control only
  Jan promises  Marie  a good person  to become
  'Jan promises Marie to become a good person.'
b. dat Jan Marie belooft een goed mens te worden.
control
  that  Jan Marie promises  a good person  to become
b'. # dat Jan Marie een goed mens belooft te worden.
subject raising
  that  Jan Marie a good person  promises  to become

      The verb dreigen in examples such as (511a) can be interpreted as a control verb or as a subject raising verb, again because there are no syntactic clues favoring one interpretation over the other; again, we abstract away from the fact that the context may disambiguate (507a) by favoring a specific interpretation.

511
a. De gemeente dreigt het kraakpand te slopen.
ambiguous
  the municipality  threatens  the squat  to demolish
  'The municipality threatens to demolish the squat.'
b. De gemeentei dreigt [PROi het kraakpand te slopen].
control
b'. De gemeentei dreigt [ti het kraakpand te slopen].
subject raising

Like example (507a) with beloven, example (511a) can be disambiguated by adding the complementizer om or by adding an anticipatory pronominal element, which surfaces here as the PP ermee'with it'; both options exclude the subject raising reading. For convenience, the elements that originate within the infinitival clause are again italicized in (512).

512
a. De gemeente dreigt om het kraakpand te slopen.
control only
  the municipality  threatens  comp  the squat  to demolish
b. De gemeente dreigt ermee het kraakpand te slopen.
control only
  the municipality  threatens  with.it  the squat  to demolish

The examples are also disambiguated when they are used as embedded clauses: if the infinitival clause is in extraposed position, as in (513a), we are normally dealing with a control structure; if we find clause splitting, as in (513b), the subject raising reading is preferred (we return to this issue in Section 5.2.2.3).

513
a. dat de gemeente dreigt het kraakpand te slopen.
control only
  that  the municipality  threatens  the squat  to demolish
  'that the municipality threatens to demolish the squat.'
b. dat de gemeente het kraakpand dreigt te slopen.
subject raising
  that the municipality  the squat  threatens  to demolish
  'that the municipality threatens to demolish the squat.'

An alternative way of disambiguating example (511a) is passivization. In the control construction the nominative subject is an agentive argument of the verb dreigen and, consequently, we expect impersonal passivization of this verb to be possible in the control reading; example (514a) shows that this expectation is indeed borne out. In the subject raising construction the nominative subject is an argument of the infinitival verb, and we expect passivization to result in promotion of the object of the infinitival verb to subject, with the concomitant suppression of the nominative subject of the corresponding active construction; example (514b) shows that this expectation is again borne out.

514
a. Er werd gedreigd het kraakpand te slopen.
control only
  there  was  threatened  the squat  to demolish
  'They threatened to demolish the squat.'
b. Het kraakpand dreigde gesloopt te worden.
subject raising only
  the squat  threatened  demolished  to be
  'The squat was in danger of being demolished.'

      Although we would in principle expect the same passivization possibilities for beloven, we have not been able to construct examples of the type in (514b) with it. This is clearly related to the fact noted earlier that examples such as (515a) cannot be construed as subject raising constructions.

515
a. De gemeente belooft het kraakpand te slopen.
control only
  the municipality  promises  the squat  to demolish
  'The municipality promises to demolish the squat.'
b. De gemeentei belooft [PROi het kraakpand te slopen].
control
  the municipality  promises  the squat  to demolish
b'. * De gemeentei belooft [ti het kraakpand te slopen].
subject raising
  the municipality  promises  the squat  to demolish

The fact that (515a) does not allow a subject raising reading correctly predicts that passivization of the verb beloven is possible, but that passivization of the infinitival verb is impossible. The former is due to the fact that the implied PRO-subject in the resulting structure in (516a) can be controlled by the noun phrase in the agentive door-phrase (which can of course also be left implicit, in which case PRO receives an arbitrary interpretation). Giving an explanation for the latter is somewhat more complicated given that we must take into account two different structures. First, the control structure in (516b) is probably excluded because the noun phrase de gemeente'the municipality' is no suitable antecedent for the implied PRO-subject of the passive infinitival clause for semantic reasons. Second, the subject raising construction in (516b') is, of course, excluded because beloven simply does not allow subject raising; cf. (515b').

516
a. Er werd door de gemeentei beloofd [PROi het kraakpand te slopen].
  there  was  by the municipality  promised the squat  to demolish
  'It was promised by the municipality to demolish the squat.'
b. * De gemeentei belooft [PRO? gesloopt te worden].
control
  the municipality  promises  demolished  to be
b'. * Het kraakpandi belooft [ti gesloopt te worden].
subject raising
  the squat  promises  demolished  to be

In short, since in the vast majority of cases the modal verb beloven takes an infinitival copular construction as its complement, and copular constructions do not allow passivization, we predict that subject raising constructions with embedded infinitival passive clauses will be rare (if existing at all).

[+]  C.  Other verbs

The subject raising verbs discussed in Subsections A and B are the ones that are common in colloquial speech. There are, however, a number of other verbs occurring in subject raising(-like) constructions that belong to the formal register, and which may be considered somewhat obsolete. Clear examples of such constructions are those with the modal verbs dunken'to deem/be of the opinion', toeschijnen'to seem', voorkomen'to appear' mentioned in (470b), which all have more or less the same meaning and behavior as the verb lijken'to appear'; for example, these verbs can all be combined with an experiencer object. The modal verb heten'to be reported', which was also listed in (470b), is more like the verb schijnen; it refers to hearsay/rumors and is thus not compatible with an experiencer object. Since constructions with such verbs do not seem to provide any new syntactic insights, we will not discuss them here.
      We have seen that subject raising verbs are characterized by the fact that they take a transparent infinitival clause. It must, however, be emphasized that the selection of a transparent infinitival clause is a necessary but not a sufficient condition for assuming subject raising; Section 5.2.2.3 will show that there are many control verbs that also allow the IPP-effect and verb clustering. Consider for instance the verb proberen in (517), which can take either an opaque or a transparent infinitival clause.

517
a. Jan heeft geprobeerd dat boek te lezen.
opaque: no IPP & extraposition
  Jan has  tried  that book  to read
  'Jan has tried to read that book.'
b. Jani heeft dat boek proberen te lezen.
transparent: IPP & verb clustering
  Jan  has  that book  try  to read
  'Jan has tried to read that book.'

That we are not dealing with subject raising in (517b) is suggested by the fact that the two constructions do not seem to differ in meaning, and is also supported by the fact illustrated in the examples in (518) that the italicized part can be pronominalized in both constructions while leaving the nominative subject Jan in place. Note in passing that in the case of (517b) pronominalization requires that the infinitival form proberen be replaced by its participial counterpart geprobeerd; this is due to the fact that there is no longer a dependent te-infinitival clause present which may trigger the IPP-effect; (518a) can therefore be seen as the pronominalized counterpart of both examples in (517).

518
a. Jan heeft dat geprobeerd.
  Jan  has  that  tried
b. * Dat heeft geprobeerd.
  that  has  tried

There are many verbs like proberen'to try' that can select either an opaque or a transparent te-infinitival clause but it is often difficult to establish for such verbs whether a subject raising analysis is possible. The problem is augmented by the earlier noted fact that many of the constructions that may be eligible for a subject raising analysis are characteristic of the formal register. In fact, it is not uncommon to find subject raising-like constructions in the formal register with atypical properties, even with subject raising verbs that occur frequently in colloquial speech. This was already indicated in Table (482), which shows that it is not impossible to find perfect-tense constructions with the verbs blijken and lijken that do not exhibit the IPP-effect. Likewise, Subsection B has shown that the subject raising verb dreigen normally does not allow extraposition, but Vliegen (2006) nevertheless found a number of (mostly formal) examples in his corpus in which the infinitival clause clearly is in extraposed position. It is not unlikely that such examples are relics from older stages of the language that have survived in the formal register.
      The above means that it often needs subtle argumentation to establish whether or not a specific verb may occur in a subject raising construction. We illustrate this with two examples. The first example involves the verb menen. Haeseryn et al. (1997:951) show that the clause selected by this verb can either be a transparent or an opaque infinitival, which show a subtle meaning difference: in (519a) the verb menen is claimed to mean "to be of the opinion", whereas in (519b) it is claimed to have the epistemic-like modal meaning "to wrongly suppose".

519
a. dat hij meent/heeft gemeend de waarheid te vertellen.
opaque
  that  he  thinks/has  thoughtPart. the truth  to tell
  'that he thinks/thought that he is/was telling the truth.'
b. dat hij de waarheid meent/heeft menen te vertellen.
transparent
  that  he  the truth  thinks/has  thinkInf.  to tell
  'that he (wrongly) believes/believed that he is/was telling the truth.'

The fact that the syntactic differences between the two examples go hand-in-hand with a difference in meaning suggests that the two constructions may require a control and a subject raising analysis, respectively. Additional evidence is, however, hard to come by. Pronominalization of the infinitival clause in (519b), for example, cannot affect the nominative subject of the clause: cf. *Dat meende. Furthermore, the nominative subject of sentence (519b) is typically animate, which suggests that the subject must satisfy selection restrictions imposed by the verb menen, which, in turn, implies that it functions as an argument of this verb and that we are therefore dealing with a control construction. Assuming that (519b) is not a subject raising but a control construction would also account for the acceptability contrast between the two primed examples in (520), in which the embedded infinitival clause is passivized.

520
a. dat Jan zijn dochter meende te hebben gezien.
transparent
  that  Jan his daughter  thought  to have  seen
  'that Jan believed to have seen his daughter.'
a'. dat zijn dochter meende te zijn gezien.
  that  his daughter  thought  to  have.been  seen
  'that his daughter thought to have been seen.'
b. dat Jan zijn gestolen auto meende te hebben gezien.
transparent
  that  Jan his stolen car  thought  to have  seen
  'that Jan believed to have seen his stolen car.'
b'. * dat zijn gestolen auto meende te zijn gezien.
  that  his stolen car  thought  to have.been  seen

The contrast between the primed passive examples in (520) can be accounted for elegantly by the control analysis in the (a)-examples in (521), which are given in main clause order for simplicity. Whereas the animate subject zijn dochter'his daughter' in (520a') satisfies the selection restrictions of menen, the inanimate subject zijn gestolen auto'his stolen car' in (520b') does not, which leads to a semantically infelicitous result: so the representation in (521a') is correctly predicted to be ungrammatical. Under the subject raising analysis in the (b)-examples, on the other hand, we cannot appeal to these selection restrictions as the surface subject is not the external argument of menen but the internal argument of the main verb of the infinitival clause. Consequently, the contrast would remain unexplained, regardless of whether the representation is deemed grammatical or ungrammatical. This strongly suggests that a subject raising analysis is not viable for constructions in which the verb menen takes a transparent infinitival complement. From this we can safely conclude that the subject raising analysis in the (b)-examples cannot be correct.

521
a. Zijn dochteri meende [PROi gezien te zijn].
control analysis
a'. * Zijn gestolen autoi meende [PROi gezien te zijn].
b. Zijn dochteri meende [ti gezien te zijn].
subject raising analysis
b'. Zijn gestolen autoi meende [ti gezien te zijn].

This is clearly different for another verb that has been claimed to belong to the set of evidential modal verbs, plegen'to be accustomed/tend'; see Vliegen (to appear). Normally speaking, subjects of constructions with this modal verb can be inanimate, as shown by Die klok pleegt achter te lopen'That clock tends to be slow' taken from the electronic Van Dale dictionary Dutch/English 2009. Another illustration of this fact is given in (522b) by means of a passive construction that is comparable to (520b').

522
a. dat wij onze computers jaarlijks plegen te controleren.
  that  we  our computers  annually  are.accustomed  to check
  'that we normally check our computers annually.'
b. dat onze computers jaarlijks gecontroleerd plegen te worden.
  that  our computers  annually  checked  are.accustomed  to be
  'that our computers are normally checked annually.'

The fact that the subject of plegen can be inanimate strongly favors the subject raising analysis in (523b): the control analysis in (523a) would again lead us to expect (522b) to evoke a violation of the selection restrictions of plegen, which requires an agentive subject when it denotes an event.

523
a. * Onze computersi plegen [PROi jaarlijks gecontroleerd te worden].
b. Onze computersi plegen [ti jaarlijks gecontroleerd te worden].

A number of other verbs from the formal register that pass the inanimacy test are the verbs behoren'to be supposed', dienen'must', hoeven'need', which are again modal in nature. We simply illustrate this by means of passive examples in (524).

524
a. Deze klok behoort/dient dagelijks opgewonden te worden.
  this clock  is.supposed/must  daily  up-wound  to be
  'This clock is supposed to/must be wound up daily.'
b. Deze klok hoeft niet dagelijks opgewonden te worden.
  this clock  need  not  daily  up-wound  to be
  'This clock need not be wound up daily.'
[+]  III.  Passive subject raising construction

We conclude the discussion of subject raising by investigating a construction of a more limited type, which we will refer to as the passive subject raising construction, subsection A starts with a discussion of the prototypical examples in (525) that involve the verbs achten'to expect' and veronderstellen'to suppose', which are often claimed to be restricted to specific registers or even to be idiomatic; cf. the lemma achten in the electronic Van Dale dictionary Dutch/English 2009.

525
Jan wordt geacht/verondersteld dat boek te lezen.
  Jan is  expected/supposed  that book  to read
'Jan is expected/supposed to read that book.'

Subsection B discusses a second set of passive subject raising constructions involving subject control verbs of the type beweren'to claim', which were characterized as obligatory control verbs in Section 5.2.2.1, sub I. Examples such as (526a) elicit different acceptability judgments from speakers; see, e.g., Bennis & Hoekstra (1989c:177) and Sturm (1990:278). They are, however, better than corresponding examples such as (526b) with non-obligatory subject control verbs like proberen'to try'.

526
a. % Jan wordt beweerd dat boek te lezen.
  Jan is  claimed  that book  to read
  'Jan is claimed to have read that book.'
b. * Jan wordt geprobeerd dat boek te lezen.
  Jan is  tried  that book  to read

Subsection C is slightly more theoretical in nature and tries to relate the contrast between the two examples in (526) to another difference that was hypothesized in Section 5.2.2.1, sub IV, to exist between obligatory and non-obligatory control verbs. The markedness of (526) does not follow from this difference, but can be assumed to reflect a typical property of semi-transparent te-infinitival clauses; see Section 5.2.2.3 for an extensive discussion of the distinction between opaque and (semi-)transparent infinitivals. For completeness' sake, we refer to Bennis & Hoekstra (1989c: Section 6.6) for an alternative proposal that departs from a different set of assumptions, but which seems quite similar in spirit at a somewhat deeper level to the one proposed here.

[+]  A.  Idiomatic passive subject raising constructions: Geacht/verondersteld worden

Typical instantiations of the passive subject raising construction are given in the primeless examples in (527); these examples are characterized by the fact that the matrix predicates are in the passive voice, without there being active counterparts; the active sentences in the primed examples are unacceptable.

527
a. Jan wordt geacht dat boek te lezen.
  Jan  is  expected  that book  to read
  'Jan is expected to read that book.'
a'. * Wij achten Jan dat boek te lezen.
  we  expect  Jan that book  to read
  Intended reading: 'We expect Jan to read that book.'
b. Jan wordt verondersteld dat boek te lezen.
  Jan is  supposed  that book  to read
  'Jan is supposed to read that book.'
b'. * We veronderstellen Jan dat boek te lezen.
  we  suppose  Jan that book  to read
  Intended reading: 'We suppose that Jan will read that book.'

It seems unlikely that the primeless examples in (527) are control constructions. The reason is that at least the verb veronderstellen is not a ditransitive verb. It is clear from (528a) that this verb does not allow an object when it takes a finite complement. Consequently, the corresponding passive construction in (528b) is impersonal: replacing the expletive pronoun er by a referential noun phrase such as Marie leads to ungrammaticality. Unfortunately, we cannot show the same for achten since this verb does not take finite argument clauses.

528
a. We veronderstellen (*Marie) [dat Jan dat boek leest].
  we  suppose     Marie   that  Jan that book  reads
  'We suppose that Jan is reading that book.'
b. Er/*Marie wordt verondersteld [dat Jan dat boek leest].
  there/Marie  is  supposed   that  Jan that book  reads
  'It is supposed that Jan is reading that book.'

The discussion of the examples in (528) implies that the nominative subject of the passive construction in (527b) cannot be an argument of veronderstellen either, but must originate within its complement clause. This implies subject raising and precludes a control analysis; see also Bennis & Hoekstra (1989c:176ff.). If we assume that the conclusion for veronderstellen carries over to the verb achten, we end up with the primeless structures in (529); the primed structures are not possible.

529
a. Jani wordt geacht [ti dat boek te lezen].
a'. * Jani wordt geacht [PROi dat boek te lezen].
b. Jani wordt verondersteld [ti dat boek te lezen].
b'. * Jani wordt verondersteld [PROi dat boek te lezen].

An empirical argument in favor of the subject raising analysis is provided by Den Besten (1985:fn.8), who shows that nominative subjects may follow the participles geacht and verondersteld in clause-final position if they are indefinite; we illustrate this in (530), in which we have italicized the subjects of the constructions for convenience. Similar examples are easy to find on the internet by means of the search string [er wordt geacht/verondersteld (g)een].

530
a. Er wordt geacht geen verschil tussen man en vrouw te zijn.
  there is supposed  no difference  between man and woman  to be
  'There is assumed to be no difference between man and woman.'
b. Er wordt verondersteld een gezagsverhouding aanwezig te zijn indien ...
  there is assumed  a power.relationship  present  to be  if
  'There is assumed to exist a relation of power if …'

If indefinite subjects can remain in their base position (see Section N8.1.4), the examples in (530) would support the claim that the nominative subjects of the constructions are base-generated in their infinitival complement clauses. Note for completeness' sake that indefinite subjects of expletive passive constructions normally precede the passivized verb in clause-final position: cf. Er is gisteren <een man> vermoord <*een man>'There was a man killed yesterday'.
      There are, however, also obvious problems for the subject raising analysis. First, it leaves unexplained why the primed examples in (527) are unacceptable: why is it impossible for the active verbs achten and veronderstellen to assign accusative case to the subject of the te-infinitivals, as is normally assumed to be possible in the corresponding English translation, which are fully grammatical? Just as surprising is the fact that the primed examples are also unacceptable if we replace the noun phrase Jan by the implied subject PRO; given that there is a suitable controller available there is no a priori reason why PRO should be excluded.

531
a. * Wij achten [Jan/PRO dat boek te lezen].
  we  expect   Jan/PRO  that book  to read
  Intended reading: 'We expect Jan/PRO to read that book.'
b. * We veronderstellen [Jan/PRO dat boek te lezen].
  we  suppose   Jan/PRO  that book  to read
  Intended reading: 'We suppose Jan/PRO to read that book.'

A second problem for assuming that the primeless examples in (527) are subject raising constructions is that Subsection II has established that subject raising normally requires verb clustering in clause-final position. The examples in (532) show, however, that the infinitival clause is normally extraposed in passive subject raising constructions; the infinitival clause normally follows the participle in clause-final position and splitting the te-infinitival gives rise to a marked result at best.

532
a. Jan wordt <?dat boek> geacht <dat boek> te lezen.
  Jan is      that book  expected  to read
  'Jan is expected to read that book.'
b. Jan wordt <??dat boek> verondersteld <dat boek> te lezen.
  Jan is        that book>   supposed  to read
  'Jan is supposed to read that book.'

This atypical behavior of the passive subject raising construction may be due to the fact that it is not part of core grammar but of the periphery (the consciously learned part) of the grammar. Seuren & Hamans (2009:fn.18), for example, claim that passive subject raising constructions are restricted to "the higher social register" and that they are not productive: they mainly occur with the predicates geacht worden and verondersteld worden; see also Den Besten (1985), who characterized even the constructions with these predicates as marked. If we are indeed dealing with a peripheral construction, it may be that its exceptional behavior is simply a reflex of the diachronic origin of the construction; see also the discussion in Subsection IIC.

[+]  B.  Passivized subject control verbs

Some speakers allow a wider variety of predicates, as is clear from the fact that the predicates in (533b) are explicitly cited in the literature as possible in passive subject raising constructions; see, e.g., Bennis & Hoekstra (1989c) and Vanden Wyngaerd (1994). Genuine examples of this type can also be found on the internet by means of the search string [wordt Vparticiple te], although one must be aware that there are also many cases that look like the products of machine translation. If the predicates in (533b) are indeed acceptable in passive subject raising constructions, the productivity of the construction may be much higher than suggested by Seuren & Hamans (2009).

533
a. Geacht worden'be expected', verondersteld worden'be supposed'
b. Aangenomen worden'be assumed', beweerd worden'be claimed', gezegd worden'be said', verwacht worden'be expected'

It should be noted, however, that it is not a priori the case that the passive predicates in (533a) and (533b) can be treated on a par, given that they differ in a non-trivial way; whereas we have seen that the former do not have any active counterpart, examples (534a&b) show that the latter correspond to active subject control constructions. For the moment we ignore the unacceptable impersonal passive example in (534c), but we will return to it shortly.

534
a. Jani beweert [PROi de beste te zijn].
subject control
  Jan  claims  the best  to be
  'Jan claims to be the best.'
b. % Jani wordt beweerd [ti de beste te zijn].
subject raising
  Jan  is  claimed  the best  to be
  'Jan is claimed to be the best.'
c. * Er wordt beweerd [PRO? de beste te zijn].
impersonal passive
  there is claimed  the best  to be

      Subject raising with passivized subject control verbs seems possible only if we are dealing with obligatory control in the sense defined in (535); See Section 5.2.2.1 for an extensive discussion of this notion.

535
Obligatory control requires the antecedent of PRO to:
a. be overtly realized in the sentence containing PRO;
b. be local (a co-argument of the infinitival clause containing PRO);
c. be a c-commanding nominal argument (subject or object);
d. be unique (cannot be "split").

Constructions with the subject control verb beweren'to claim' satisfy the definition in (535), as is clear from the fact illustrated in example (534c) that they do not allow impersonal passivization. The subject control verb proberen'to try' in (536a), on the other hand, does not involve obligatory control, as is clear from the fact illustrated in (536c) that it allows impersonal passivization. That proberen cannot be found in the passive subject raising construction is clear from the fact that (536b) is rejected by all speakers.

536
a. Jani probeerde [(om) PROi de deur te sluiten].
subject control
  Jan  tried  comp  the door  to close
  'Jan tried to close the door.'
b. * Jani werd geprobeerd [(om) ti de deur te sluiten].
subject raising
  Jan  was  tried   comp the door  to close
c. Er werd geprobeerd [(om) PROarb de deur te sluiten].
imp. passive
  there  was   tried   comp the door  to close
  'It was tried to close the door.'
[+]  C.  Obligatory versus non-obligatory control verbs

This subsection shows that the mutual exclusion of the (b)- and (c)-examples in (534) and (536) can be accounted for by appealing to the distinction between om + te- and te-infinitivals made in Section 5.2.2.1, sub IV. The hypothesis formulated there was that om + te-infinitivals are CPs and that CP-boundaries block locally restricted syntactic dependencies like NP-movement (e.g., subject raising) and obligatory control, whereas te-infinitivals are TPs and TP-boundaries do not block such dependencies. We repeat this cluster of hypotheses here as (537), which in tandem express that locally restricted syntactic dependencies can be established across the boundary of a te-, but not across the boundary of an om + te-infinitival.

537
a. Hypothesis I: om + te-infinitivals are CPs.
b. Hypothesis II: te-infinitivals are TPs.
c. Hypothesis III: CPs constitute islands for syntactic dependencies.
b. Hypothesis IV: TPs do not constitute islands for syntactic dependencies.

The claim that CPs but not TPs are syntactic islands for obligatory control was used to account for the fact that verbs like beweren, which select te-infinitivals as their complement, trigger obligatory control, while verbs like proberen, which select om + te-infinitivals, involve non-obligatory control.

538
a. Jani beweert [TP PROi de beste te zijn].
obligatory control
  Jan  claims  the best  to be
b. Jani probeerde [CP (om) PROi de deur te sluiten].
non-obligatory control
  Jan  tried  comp  the door  to close

This difference between beweren and proberen is confirmed by the difference in behavior with respect to impersonal passivization illustrated in (534c) and (536c), which are repeated here in a slightly more precise form as (539). In accordance with hypothesis III, the CP-complement of proberen does not allow the PRO-subject to enter into an obligatory control relation with a controller in the matrix clause, and (535a) therefore allows the PRO-subject in (539b) to be controlled by the implied agent of the matrix clause, and thus to be assigned arbitrary reference. In accordance with hypothesis IV, the TP-complement of beweren does allow the PRO-subject to enter into an obligatory control relation with a controller in the matrix clause, and consequently (535a) prohibits control of the PRO-subject in (539a) by the implied agent of the matrix clause; consequently, PRO cannot be assigned an interpretation and the structure is uninterpretable.

539
a. * Er wordt beweerd [TP PRO? de beste te zijn].
  there is claimed  the best  to be
b. Er werd geprobeerd [CP (om) PROarb de deur te sluiten].
  there  was   tried  comp the door  to close
  'It was tried to close the door.'

Interestingly, the difference in behavior with respect to subject raising illustrated in (534b) and (536b), repeated here in a slightly more precise form as (540), follows from the same set of assumptions without further ado. According to hypothesis IV, the TP-complement of beweren does not block movement and thus allows the subject raising structure in (540a). Hypothesis IV, on the other hand, predicts that the CP-complement of proberen does block movement and hence excludes the structure in (540b).

540
a. % Jani wordt beweerd [TPti de beste te zijn].
  Jan  is  claimed  the best  to be
  'Jan is claimed to be the best.'
b. * Jani werd geprobeerd [CP (om) ti de deur te sluiten].
  Jan  was  tried  comp the door  to close

The hypotheses I-IV thus only leave us with the question as to why subject raising in examples such as (540a) is considered marked by most speakers, given that this clearly does not follow from the discussion above. We believe that this can be related to the fact that passive subject raising constructions involve extraction from a te-infinitival in extraposed position; Section 5.2.2.3, sub VII, will show on the basis of independent empirical evidence that movement from such infinitival clauses is more generally judged to be acceptable, but marked.

References:
  • Barbiers, Sjef2006The syntax of modal auxiliariesEveraert, Martin & Riemdijk, Henk van (eds.)The Blackwell companion to syntax5Malden, Ma/OxfordBlackwell Publishing1-22
  • Bennis, Hans & Hoekstra, Teun1989PRO and the Binding TheoryBennis, Hans & Kemenade, Ans van (eds.)Linguistics in the Netherlands 1989Dordrecht11-20
  • Bennis, Hans & Hoekstra, Teun1989Generatieve grammaticaDordrechtForis Publications
  • Bennis, Hans & Hoekstra, Teun1989Generatieve grammaticaDordrechtForis Publications
  • Bennis, Hans & Hoekstra, Teun1989Generatieve grammaticaDordrechtForis Publications
  • Bennis, Hans & Hoekstra, Teun1989Generatieve grammaticaDordrechtForis Publications
  • Bennis, Hans & Hoekstra, Teun1989Generatieve grammaticaDordrechtForis Publications
  • Bennis, Hans & Hoekstra, Teun1989Generatieve grammaticaDordrechtForis Publications
  • Besten, Hans den1985The ergative hypothesis and free word order in Dutch and GermanToman, Jindřich (ed.)Studies in German GrammarDordrecht/CinnaminsonForis Publications23-65
  • Besten, Hans den1985The ergative hypothesis and free word order in Dutch and GermanToman, Jindřich (ed.)Studies in German GrammarDordrecht/CinnaminsonForis Publications23-65
  • Bruggen, Femmy van1980/1Schijnen, lijken, blijkenTabu1154-63
  • Dreumel, Simon van & Coppen, Peter-Arno2003Surface analysis of the verbal cluster in DutchLinguistics4151--81
  • Grice, H.P1975Logic and conversationCole, P. & Morgan, J. (eds.)Speech acts: Syntax and Semantics 3New YorkAcademic Press41-58
  • Haegeman, Liliane2006Clitic Climbing and the dual status of <i>sembrare</i>Linguistic Inquiry37484-501
  • Haeseryn, Walter, Romijn, Kirsten, Geerts, Guido, Rooij, Jaap de & Toorn, Maarten C. van den1997Algemene Nederlandse spraakkunstGroningenNijhoff
  • Haeseryn, Walter, Romijn, Kirsten, Geerts, Guido, Rooij, Jaap de & Toorn, Maarten C. van den1997Algemene Nederlandse spraakkunstGroningenNijhoff
  • Haeseryn, Walter, Romijn, Kirsten, Geerts, Guido, Rooij, Jaap de & Toorn, Maarten C. van den1997Algemene Nederlandse spraakkunstGroningenNijhoff
  • Horst, Joop van der2008Geschiedenis van de Nederlandse syntaxisLeuvenUniversitaire Pers Leuven
  • Horst, Joop van der2008Geschiedenis van de Nederlandse syntaxisLeuvenUniversitaire Pers Leuven
  • Koring, Loes2013Seemingly similar. Subjects and displacement in grammar, processing, and acquisitionUniversity of UtrechtThesis
  • Koster, Jan1984Anaphoric and non-anaphoric controlLinguistic Inquiry15417-459
  • Palmer, F.R2001Mood and ModalityCambridge University Press
  • Reuland, Eric1983Government and the search for <i>aux</i>es, Vol. 1Heny, Frank & Richards, Barry (eds.)Linguistic categories: auxiliaries and related puzzlesDordrechtReidel99-168
  • Rutten, Jean-Baptist1991Infinitival complements and auxiliariesAmsterdamUniversity of AmsterdamThesis
  • Sanders, José & Spooren, Wilbert1996Subjectivity and certainty in epistemic modality: A study of Dutch epistemic modifiersCognitive Linguistics7241-264
  • Schmid, Tanja2005Infinitival syntax. Infinitivus Pro Participio as a repair strategyAmsterdam/PhiladelphiaJohn Benjamins Publishing Company
  • Seuren, Pieter & Hamans, Camiel2009Semantic conditions on syntactic rules: evidentiality and auxiliation in English and DutchFolia Linguistica43135-169
  • Seuren, Pieter & Hamans, Camiel2009Semantic conditions on syntactic rules: evidentiality and auxiliation in English and DutchFolia Linguistica43135-169
  • Sturm, Arie1990HerschrijvenForum der Letteren31271-290
  • Verhagen, Arie2005Constructions of intersubjectivity: discourse, syntax, and cognitionOxford/New YorkOxford University Press
  • Verhagen, Arie2005Constructions of intersubjectivity: discourse, syntax, and cognitionOxford/New YorkOxford University Press
  • Vliegen, Maurice2010Lexikalische Evidentialität: das niederländische <i>blijken</i>. Ein evidenter Fall der GrammatikalisierungCate, Abraham ten, Rapp, Reinhard, Strässler, Jurg, Vliegen, Maurice & Weber, Heinrich (eds.)Grammatik Praxis GeschichteTübingenNarr209-216
  • Vliegen, Maurice2011Evidentiality. Dutch <i>seem </i>and <i>appear </i>verbs: <i>blijken</i>, <i>lijken</i>, <i>schijnen</i>Nouwen, Rick & Elenbaas, Marion (eds.)Linguistics in the Netherlands 2011Amsterdam/Philadelphia125-137
  • Vliegen, Maurice2011Evidentiality. Dutch <i>seem </i>and <i>appear </i>verbs: <i>blijken</i>, <i>lijken</i>, <i>schijnen</i>Nouwen, Rick & Elenbaas, Marion (eds.)Linguistics in the Netherlands 2011Amsterdam/Philadelphia125-137
  • Vliegen, Maurice2011Evidentiality. Dutch <i>seem </i>and <i>appear </i>verbs: <i>blijken</i>, <i>lijken</i>, <i>schijnen</i>Nouwen, Rick & Elenbaas, Marion (eds.)Linguistics in the Netherlands 2011Amsterdam/Philadelphia125-137
  • Vliegen, Maurice, Cornillie, Bert & Delbecque, Nicole2006The relation between lexical and epistemic readings: The equivalents of <i>promise </i>and <i>threaten </i>in Dutch and GermanBelgian Journal of Linguistics20 [Topics in subjectification and modalization, edited by Bert Cornillie & Nicole Delbecque]73-95
  • Vliegen, Maurice, Cornillie, Bert & Delbecque, Nicole2006The relation between lexical and epistemic readings: The equivalents of <i>promise </i>and <i>threaten </i>in Dutch and GermanBelgian Journal of Linguistics20 [Topics in subjectification and modalization, edited by Bert Cornillie & Nicole Delbecque]73-95
  • Vliegen, Maurice, Cornillie, Bert & Delbecque, Nicole2006The relation between lexical and epistemic readings: The equivalents of <i>promise </i>and <i>threaten </i>in Dutch and GermanBelgian Journal of Linguistics20 [Topics in subjectification and modalization, edited by Bert Cornillie & Nicole Delbecque]73-95
  • Vliegen, Maurice, Cornillie, Bert & Delbecque, Nicole2006The relation between lexical and epistemic readings: The equivalents of <i>promise </i>and <i>threaten </i>in Dutch and GermanBelgian Journal of Linguistics20 [Topics in subjectification and modalization, edited by Bert Cornillie & Nicole Delbecque]73-95
  • Vliegen, Maurice & Tóth, Josefto appearDas niederländische <i>blijken</i> samt seinen lexikalischen Pendants im Deutschen. Ein kontrastives ProblemTóth, Josef (ed.)Akten des 45. Linguistischen Kolloquiums. 16. bis 18. September 2010HungaryUniversity of Veszprém
  • Wyngaerd, Guido vanden1994PRO-legomena. Distribution and Reference of infinitival subjectsLinguistic Models 19Berlin/New YorkMouton de Gruyter
report errorprintcite